© 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. 13799
    Prev
    Next
    INDEX
    Explanations

    descriptions of natural landscapes and ecological events

    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
    220
    +0.07
    0.5%
    622
    +0.06
    0.5%
    150
    +0.05
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    235
    +0.07
    0.11
    639
    +0.06
    0.09
    97
    +0.05
    0.04
    Negative Logits
    rador
    -0.50
     Gutenberg
    -0.48
    esson
    -0.45
    urion
    -0.42
    ICLE
    -0.42
    ector
    -0.41
    ensable
    -0.39
    slot
    -0.39
    ixel
    -0.39
    urated
    -0.39
    POSITIVE LOGITS
     greens
    0.48
    theless
    0.43
     centr
    0.41
    clad
    0.40
     professions
    0.38
     compounds
    0.38
    borg
    0.38
     tones
    0.37
     dinosaurs
    0.36
    kees
    0.36
    Activations Density 11.108%

    No Known Activations