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 E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 13867
    Prev
    Next
    INDEX
    Explanations

    financial transactions and debts

    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
    352
    +0.18
    0.6%
    385
    +0.17
    0.5%
    739
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.18
    0.10
    352
    +0.17
    0.07
    195
    +0.15
    0.02
    Negative Logits
    =-=-=-=-=-=-=-=-
    -1.07
    anooga
    -1.04
     Airl
    -1.01
    unch
    -0.95
    iq
    -0.95
     Hobby
    -0.93
    nets
    -0.92
    ulton
    -0.91
    nel
    -0.91
     Expand
    -0.91
    POSITIVE LOGITS
     owed
    1.52
     repay
    1.35
     repayment
    1.33
     forgiveness
    1.30
     forgiven
    1.29
     repaid
    1.21
     debts
    1.20
     borrowers
    1.11
     creditors
    1.11
    owship
    1.10
    Activations Density 1.434%

    No Known Activations