Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 4548
    Prev
    Next
    INDEX
    Explanations

    sports-related terms and statistics in the context of various games and players

    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
    314
    +0.08
    0.6%
    674
    +0.06
    0.4%
    320
    +0.06
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    120
    +0.08
    0.11
    314
    +0.06
    -0.11
    100
    +0.06
    0.08
    Negative Logits
    hunt
    -0.53
    enegger
    -0.46
     Galile
    -0.46
    drm
    -0.43
     breathing
    -0.43
    fman
    -0.43
    beard
    -0.42
    shore
    -0.42
    hawk
    -0.42
    tle
    -0.41
    POSITIVE LOGITS
    Ö¼
    0.60
    ÃĽ
    0.44
     variance
    0.40
    llah
    0.38
    apest
    0.37
    ा
    0.37
    ccording
    0.37
    vertisement
    0.37
     Adinida
    0.36
    Specific
    0.35
    Activations Density 17.467%

    No Known Activations