© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 35930
    Prev
    Next
    INDEX
    Explanations

    specific actions or events in a narrative context, particularly focusing on characters' movements and interactions

    oai_token-act-pair · gpt-3.5-turbo

    action-related verbs and phrases that convey a sense of movement or activity

    oai_token-act-pair · gpt-3.5-turboTriggered by @danbraun
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.18
    1.0%
    560
    +0.11
    0.6%
    326
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.18
    0.02
    70
    +0.11
    0.13
    665
    +0.08
    0.15
    Negative Logits
     Kardash
    -0.76
    modern
    -0.60
     Persons
    -0.59
    milo
    -0.58
    actual
    -0.57
    DragonMagazine
    -0.57
     dudes
    -0.57
     predecessors
    -0.57
    interstitial
    -0.57
     Defendants
    -0.56
    POSITIVE LOGITS
     angrily
    0.81
     nervously
    0.77
     softly
    0.76
     apolog
    0.73
     calmly
    0.72
     impatient
    0.71
     sob
    0.70
     cautiously
    0.69
     gently
    0.69
    rawdownloadcloneembedreportprint
    0.69
    Activations Density 10.222%

    No Known Activations