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. 45080
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names, particularly ones that contain the letter sequence "Bu"

    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
    211
    +0.12
    0.6%
    653
    +0.11
    0.5%
    343
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.14
    288
    +0.11
    0.14
    651
    +0.11
    0.09
    Negative Logits
    wear
    -0.75
     Cosponsors
    -0.70
    bucks
    -0.69
    bread
    -0.67
    Introduced
    -0.66
    */(
    -0.66
     Butterfly
    -0.66
     Nadu
    -0.65
    papers
    -0.65
     Sabb
    -0.63
    POSITIVE LOGITS
    xton
    1.12
    cci
    0.97
    selage
    0.95
    arded
    0.95
    isons
    0.94
    eno
    0.92
    ilts
    0.92
    igslist
    0.92
    zag
    0.89
    querque
    0.89
    Activations Density 3.661%

    No Known Activations