© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 14290
    Prev
    Next
    INDEX
    Explanations

    comments and opinions related to sports and sports events

    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
    760
    +0.11
    0.5%
    124
    +0.10
    0.5%
    653
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.11
    0.25
    8
    +0.10
    0.13
    14
    +0.09
    0.07
    Negative Logits
     [-
    -0.71
    andise
    -0.64
    ãģ®ç
    -0.64
    "[
    -0.62
    earch
    -0.62
    ãĤ´
    -0.61
    -[
    -0.61
    "(
    -0.60
    ".[
    -0.60
    licts
    -0.59
    POSITIVE LOGITS
     someday
    0.90
    rosso
    0.86
     somew
    0.86
     misunder
    0.74
     typo
    0.72
     quir
    0.70
    tim
    0.68
     wiser
    0.67
     inadvert
    0.67
    Äĩ
    0.66
    Activations Density 11.275%

    No Known Activations