© 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 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. 6383
    Prev
    Next
    INDEX
    Explanations

    numerical patterns within text data

    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
    52
    +0.17
    0.6%
    461
    +0.15
    0.6%
    6
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.17
    0.11
    52
    +0.15
    0.07
    6
    +0.14
    0.08
    Negative Logits
    reditary
    -0.98
    sworth
    -0.96
    acco
    -0.96
    opter
    -0.90
    matically
    -0.88
    ems
    -0.88
    entials
    -0.87
    ophe
    -0.86
    mers
    -0.86
    isters
    -0.86
    POSITIVE LOGITS
    Ñĭ
    1.12
    âĶĢ
    1.03
    Ñĥ
    1.02
    âĶĢâĶĢ
    0.98
    âĶĢâĶĢâĶĢâĶĢ
    0.95
    âĸĪ
    0.93
    а
    0.90
    INE
    0.89
    âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ
    0.89
    tenance
    0.89
    Activations Density 1.824%

    No Known Activations