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. 4278
    Prev
    Next
    INDEX
    Explanations

    the word "turn" followed by a number indicating a significant action or change

    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
    +0.22
    0.9%
    195
    +0.15
    0.6%
    128
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.22
    0.05
    70
    +0.15
    0.10
    599
    +0.11
    0.07
    Negative Logits
    è¦ļéĨĴ
    -1.05
    ropolitan
    -0.92
    capacity
    -0.82
    mma
    -0.80
    inately
    -0.78
    ording
    -0.76
    foundation
    -0.72
    bour
    -0.71
    llah
    -0.71
    lain
    -0.69
    POSITIVE LOGITS
     awa
    0.87
     into
    0.85
    coat
    0.81
     shif
    0.77
    agra
    0.76
    around
    0.76
     inward
    0.75
     crank
    0.75
    about
    0.75
     sour
    0.74
    Activations Density 8.651%

    No Known Activations