Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 13560
    Prev
    Next
    INDEX
    Explanations

    statements expressing belief, opinion, or personal perspective

    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
    +0.30
    1.4%
    483
    +0.12
    0.6%
    107
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.30
    0.28
    119
    +0.12
    0.15
    739
    +0.11
    0.06
    Negative Logits
    artney
    -0.72
    elle
    -0.67
    ption
    -0.64
    fare
    -0.62
     purportedly
    -0.61
    bies
    -0.59
    eller
    -0.58
    ueless
    -0.57
    lopp
    -0.57
    iae
    -0.57
    POSITIVE LOGITS
    poke
    0.80
     strongly
    0.71
     myself
    0.68
    ħ
    0.68
     passionately
    0.65
    ĸ
    0.63
     fortunate
    0.63
     personally
    0.60
    §
    0.60
    ļé
    0.59
    Activations Density 11.750%

    No Known Activations