© 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. 40437
    Prev
    Next
    INDEX
    Explanations

    information related to publications and books

    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
    +0.28
    1.5%
    393
    +0.20
    1.1%
    479
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.28
    0.18
    393
    +0.20
    0.29
    546
    +0.08
    0.18
    Negative Logits
    ose
    -0.62
    atsuki
    -0.61
    assed
    -0.61
    worldly
    -0.61
    weak
    -0.60
    ¬¼
    -0.59
    obin
    -0.57
    suff
    -0.57
    osing
    -0.57
    keyes
    -0.56
    POSITIVE LOGITS
     Inc
    0.96
     LLC
    0.94
     Ltd
    0.85
     etc
    0.79
    Gall
    0.75
     ISBN
    0.72
     aka
    0.72
     Suite
    0.72
     supra
    0.66
    76561
    0.66
    Activations Density 9.717%

    No Known Activations