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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4072
    Prev
    Next
    INDEX
    Explanations

    references to physical objects that serve as central or stabilizing points

    oai_token-act-pair · gpt-3.5-turbo
    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
    +0.16
    0.7%
    549
    +0.14
    0.7%
    526
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.16
    0.21
    326
    +0.14
    0.19
    145
    +0.12
    0.07
    Negative Logits
    INESS
    -0.84
    ensation
    -0.83
    BLE
    -0.76
    soDeliveryDate
    -0.70
    STEM
    -0.69
    osponsors
    -0.67
     Nicotine
    -0.67
    cigarettes
    -0.66
    DOS
    -0.66
    theless
    -0.66
    POSITIVE LOGITS
     anch
    1.03
    orage
    0.99
     anchor
    0.95
     anchored
    0.92
     anchors
    0.86
     tenant
    0.85
    ovy
    0.82
    ments
    0.81
    uckland
    0.74
    ovies
    0.74
    Activations Density 6.186%

    No Known Activations