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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 13423
    Prev
    Next
    INDEX
    Explanations

    references to the passage of time and historical context

    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
    7
    +0.09
    0.5%
    739
    +0.08
    0.5%
    268
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.09
    0.16
    55
    +0.08
    0.17
    14
    +0.08
    0.07
    Negative Logits
     tamp
    -0.57
    avier
    -0.52
     nickname
    -0.49
     tmp
    -0.49
     Favorite
    -0.48
     incidental
    -0.48
    orically
    -0.48
    uum
    -0.48
     labor
    -0.47
    iatric
    -0.47
    POSITIVE LOGITS
     ripe
    0.59
    externalActionCode
    0.57
    QUI
    0.53
     flanked
    0.52
    ope
    0.51
    cation
    0.51
    window
    0.51
    rawdownloadcloneembedreportprint
    0.51
    ste
    0.51
     surrounded
    0.50
    Activations Density 9.113%

    No Known Activations