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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 14606
    Prev
    Next
    INDEX
    Explanations

    references to titles or names of concepts and terms

    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
    647
    +0.18
    0.7%
    393
    +0.16
    0.6%
    52
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    131
    +0.18
    0.05
    647
    +0.16
    0.05
    52
    +0.11
    0.04
    Negative Logits
     istg
    -0.95
    inth
    -0.91
    jen
    -0.86
    âĢ¢âĢ¢âĢ¢âĢ¢
    -0.84
    Brew
    -0.83
     unden
    -0.82
     ensu
    -0.81
    isman
    -0.80
    -------------
    -0.79
     Commodore
    -0.79
    POSITIVE LOGITS
    phas
    0.91
    called
    0.85
     bullshit
    0.82
     "
    0.81
     '
    0.80
     `
    0.79
    call
    0.78
    anus
    0.78
    ãĢİ
    0.77
     SIG
    0.77
    Activations Density 0.409%

    No Known Activations