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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 32918
    Prev
    Next
    INDEX
    Explanations

    names of specific individuals who were drafted or involved in professional sports

    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
    554
    +0.14
    0.7%
    447
    +0.13
    0.7%
    202
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    554
    +0.14
    -0.11
    566
    +0.13
    0.12
    761
    +0.11
    0.10
    Negative Logits
    plug
    -0.71
    fal
    -0.66
    ebin
    -0.65
    fm
    -0.63
    eu
    -0.61
    fol
    -0.61
    fing
    -0.61
    urst
    -0.61
    edom
    -0.61
    usable
    -0.60
    POSITIVE LOGITS
    dinand
    0.71
    ãĥ³ãĤ¸
    0.66
     lottery
    0.65
    çīĪ
    0.65
    ãĤ´
    0.63
     picks
    0.62
    ç·
    0.61
     Avalanche
    0.61
    avorite
    0.60
     drafted
    0.59
    Activations Density 11.216%

    No Known Activations