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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 11075
    Prev
    Next
    INDEX
    Explanations

    statements about achieving success or making progress towards a goal

    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
    141
    +0.16
    0.5%
    246
    +0.13
    0.4%
    393
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    526
    +0.16
    0.00
    608
    +0.13
    0.06
    570
    +0.12
    0.02
    Negative Logits
     suscept
    -0.93
     contrace
    -0.86
     Consent
    -0.83
    arna
    -0.82
     strengths
    -0.81
     expressions
    -0.80
    rator
    -0.80
    rants
    -0.78
    ials
    -0.78
     ACTIONS
    -0.77
    POSITIVE LOGITS
    ophys
    0.97
    finder
    0.90
    éĹ
    0.89
    ozo
    0.89
    opolis
    0.89
     unsc
    0.88
    plom
    0.88
     cutoff
    0.86
     Unloaded
    0.86
    iler
    0.85
    Activations Density 1.038%

    No Known Activations