© 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. 46004
    Prev
    Next
    INDEX
    Explanations

    descriptive details and interactions among characters in a narrative

    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
    +0.34
    1.6%
    393
    +0.21
    1.0%
    589
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.34
    0.33
    36
    +0.21
    0.16
    589
    +0.12
    0.12
    Negative Logits
    ciples
    -0.77
    Ľ
    -0.76
    uously
    -0.69
    orah
    -0.68
    istics
    -0.68
    orians
    -0.66
    asters
    -0.66
    ATURES
    -0.65
    ormons
    -0.64
    oby
    -0.63
    POSITIVE LOGITS
     pesky
    1.15
     kind
    0.92
     same
    0.86
     particular
    0.85
     sort
    0.84
     fateful
    0.83
     elusive
    0.80
    cher
    0.77
     type
    0.76
     sucker
    0.72
    Activations Density 14.074%

    No Known Activations