© 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. 6056
    Prev
    Next
    INDEX
    Explanations

    phrases warning against potential negative outcomes

    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
    326
    +0.17
    0.7%
    551
    +0.14
    0.6%
    19
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    283
    +0.17
    0.02
    57
    +0.14
    0.03
    535
    +0.14
    0.02
    Negative Logits
    LINE
    -0.89
     Nationwide
    -0.87
    Cola
    -0.86
    issance
    -0.83
    scope
    -0.82
     Congratulations
    -0.81
    essor
    -0.80
    ird
    -0.78
    machine
    -0.77
    LIN
    -0.77
    POSITIVE LOGITS
     lest
    1.19
     detract
    0.94
     conce
    0.90
    00200000
    0.89
     inward
    0.89
     thou
    0.87
     anyone
    0.86
    entimes
    0.86
     otherwise
    0.86
    terness
    0.84
    Activations Density 0.399%

    No Known Activations