© 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. 40412
    Prev
    Next
    INDEX
    Explanations

    dates and time-related information

    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
    +0.20
    1.0%
    709
    +0.09
    0.5%
    646
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.20
    0.22
    640
    +0.09
    0.29
    607
    +0.09
    0.13
    Negative Logits
     Cosponsors
    -0.67
     Provided
    -0.61
    Syn
    -0.53
    ãĥĦ
    -0.53
    igham
    -0.52
    <?
    -0.51
     dy
    -0.50
    ese
    -0.49
     Zeal
    -0.49
     Blueprint
    -0.48
    POSITIVE LOGITS
     recess
    0.74
    cation
    0.65
    ences
    0.64
     onwards
    0.60
    emetery
    0.60
     onward
    0.58
    OHN
    0.57
    imental
    0.57
    ence
    0.56
    ç
    0.54
    Activations Density 14.291%

    No Known Activations