© 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. 5966
    Prev
    Next
    INDEX
    Explanations

    Pokémon names and related information

    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
    714
    +0.05
    0.4%
    549
    +0.05
    0.4%
    328
    +0.04
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    545
    +0.05
    0.01
    326
    +0.05
    0.01
    284
    +0.04
    0.01
    Negative Logits
     Penal
    -0.36
     rehe
    -0.36
     saints
    -0.34
    assed
    -0.34
     fasting
    -0.32
     booted
    -0.32
     sins
    -0.31
     subscript
    -0.31
     Saints
    -0.30
     IRA
    -0.30
    POSITIVE LOGITS
    sembly
    0.41
    atform
    0.38
    erie
    0.35
     DF
    0.35
    science
    0.34
    ãĤ´ãĥ³
    0.33
    xual
    0.33
    intend
    0.33
    date
    0.32
     Plants
    0.32
    Activations Density 0.011%

    No Known Activations