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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 34535
    Prev
    Next
    INDEX
    Explanations

    words related to self-awareness, personal development, and critical thinking

    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
    +0.24
    1.1%
    102
    +0.14
    0.6%
    676
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.24
    -0.09
    102
    +0.14
    0.22
    331
    +0.13
    0.15
    Negative Logits
    ibus
    -0.66
    iage
    -0.65
    CRIPTION
    -0.65
    endas
    -0.65
     Flavoring
    -0.63
     Sut
    -0.63
    umption
    -0.62
    ructure
    -0.60
    rities
    -0.60
     clause
    -0.60
    POSITIVE LOGITS
    compatible
    1.00
    eligible
    0.94
    theless
    0.90
    actly
    0.88
    ificantly
    0.88
    dependent
    0.84
     incarn
    0.79
    fed
    0.76
     liable
    0.75
    functional
    0.75
    Activations Density 21.628%

    No Known Activations