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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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 E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 455
    Prev
    Next
    INDEX
    Explanations

    names of individuals, specifically the name "Frank"

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    339
    +0.15
    0.5%
    55
    +0.15
    0.5%
    52
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.15
    0.09
    55
    +0.15
    0.07
    52
    +0.14
    0.05
    Negative Logits
    ILA
    -1.31
    uana
    -1.12
    ADRA
    -1.05
    kaya
    -1.01
    alid
    -1.01
    iquid
    -0.99
    halla
    -0.99
    TPPStreamerBot
    -0.97
    atform
    -0.95
    arre
    -0.95
    POSITIVE LOGITS
    furt
    1.74
    fur
    1.69
    lins
    1.37
    fort
    1.31
    enstein
    1.29
    ness
    1.17
     Geh
    1.13
    lin
    1.11
    ly
    1.04
     Ocean
    1.04
    Activations Density 2.069%

    No Known Activations