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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 13466
    Prev
    Next
    INDEX
    Explanations

    mentions of specific events and actions in narrative texts

    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
    +0.35
    1.7%
    218
    +0.12
    0.6%
    59
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.35
    0.28
    231
    +0.12
    0.14
    563
    +0.10
    0.16
    Negative Logits
    inguishable
    -0.63
    onica
    -0.62
    NRS
    -0.60
    tools
    -0.60
    optional
    -0.60
    usercontent
    -0.60
    atis
    -0.59
    ettle
    -0.59
    ang
    -0.59
    ele
    -0.58
    POSITIVE LOGITS
     periods
    0.75
     weekends
    0.72
    itialized
    0.70
    ventory
    0.70
     times
    0.66
     halftime
    0.64
     renovations
    0.63
     evenings
    0.62
     Seasons
    0.62
    flation
    0.62
    Activations Density 8.283%

    No Known Activations