© 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. 45847
    Prev
    Next
    INDEX
    Explanations

    descriptions of physical attributes, particularly height and weight

    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
    17
    +0.11
    0.5%
    122
    +0.10
    0.4%
    155
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    35
    +0.11
    0.07
    678
    +0.10
    0.07
    610
    +0.09
    0.05
    Negative Logits
     Tayyip
    -0.77
     Warn
    -0.73
     Trials
    -0.71
     Administ
    -0.70
    sth
    -0.70
    IENCE
    -0.69
    orate
    -0.69
    ostics
    -0.69
    ighting
    -0.67
    Results
    -0.66
    POSITIVE LOGITS
     tall
    0.86
     lumber
    0.83
     taller
    0.79
     floppy
    0.76
     towering
    0.76
     humanoid
    0.75
     wheelchair
    0.74
     mammoth
    0.73
     blonde
    0.73
     skinny
    0.72
    Activations Density 6.646%

    No Known Activations