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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 45134
    Prev
    Next
    INDEX
    Explanations

    specific keywords related to match coverage and events in sports contexts

    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
    +0.27
    1.1%
    607
    +0.13
    0.5%
    651
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.27
    0.16
    442
    +0.13
    0.08
    651
    +0.12
    0.10
    Negative Logits
    sonian
    -0.80
    irtual
    -0.78
    achev
    -0.74
    ħĭ
    -0.73
    é»Ĵ
    -0.71
    ible
    -0.70
    raft
    -0.70
     captcha
    -0.69
    onne
    -0.69
    umatic
    -0.68
    POSITIVE LOGITS
    making
    1.06
    maker
    1.02
    makers
    0.98
    nings
    0.92
    eer
    0.87
    stick
    0.84
    sticks
    0.83
    match
    0.83
    eering
    0.80
     ups
    0.79
    Activations Density 6.085%

    No Known Activations