© 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. 34808
    Prev
    Next
    INDEX
    Explanations

    references to sports teams and players, as well as football-specific terminology

    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
    +0.40
    2.4%
    640
    +0.15
    0.9%
    576
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.40
    0.34
    6
    +0.15
    0.21
    64
    +0.10
    -0.22
    Negative Logits
     administrator
    -0.53
    uly
    -0.51
     swinging
    -0.51
     casting
    -0.50
     veter
    -0.49
     assignment
    -0.48
    uba
    -0.48
     abl
    -0.47
     lad
    -0.47
     advant
    -0.47
    POSITIVE LOGITS
     [+
    0.75
     Beware
    0.67
     Lack
    0.63
     Conclusion
    0.60
     Recommend
    0.60
     Learns
    0.59
     Despite
    0.56
     Seriously
    0.56
    jpg
    0.56
     Compare
    0.56
    Activations Density 13.050%

    No Known Activations