© 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
    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. 32538
    Prev
    Next
    INDEX
    Explanations

    technical instructions or procedures

    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
    393
    +0.19
    0.9%
    159
    +0.14
    0.7%
    116
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.19
    0.16
    393
    +0.14
    0.24
    589
    +0.12
    0.11
    Negative Logits
    aws
    -0.68
     Papers
    -0.67
    doms
    -0.66
    aths
    -0.66
    lee
    -0.65
    pots
    -0.64
    Personal
    -0.63
    rior
    -0.63
    ashes
    -0.63
    oway
    -0.62
    POSITIVE LOGITS
     trope
    0.97
     latter
    0.91
     particular
    0.84
     week
    0.77
     article
    0.75
     sucker
    0.75
     same
    0.75
     morning
    0.73
     approximation
    0.73
     incarnation
    0.73
    Activations Density 9.525%

    No Known Activations