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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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 Local
    9. 6-RES_SCL-AJT
    11. 4164
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal proceedings and documentation

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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
    447
    +0.26
    1.2%
    393
    +0.20
    0.9%
    649
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.26
    0.28
    159
    +0.20
    0.12
    608
    +0.10
    0.19
    Negative Logits
    seless
    -0.78
    plex
    -0.77
    rex
    -0.74
    rox
    -0.71
    gal
    -0.69
    ulo
    -0.68
    gypt
    -0.68
    itals
    -0.66
    grad
    -0.66
    itia
    -0.66
    POSITIVE LOGITS
    THING
    1.28
    WHERE
    1.07
     significant
    0.92
    ONE
    0.89
     meaningful
    0.89
     particular
    0.88
    place
    0.87
    ones
    0.83
     substantive
    0.79
    body
    0.79
    Activations Density 11.958%

    No Known Activations