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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 45425
    Prev
    Next
    INDEX
    Explanations

    names or references related to "Kanan" or similar variations

    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
    659
    +0.16
    0.8%
    386
    +0.11
    0.6%
    402
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.16
    0.08
    544
    +0.11
    0.04
    386
    +0.09
    0.07
    Negative Logits
    Ö¼
    -0.76
    llor
    -0.62
     Aires
    -0.59
    suit
    -0.58
    inals
    -0.57
    inately
    -0.56
     FIGHT
    -0.56
     Discrimination
    -0.56
    é¾
    -0.56
    ×Ļ×
    -0.55
    POSITIVE LOGITS
    ji
    0.97
    hof
    0.96
    ako
    0.87
    igan
    0.85
    igans
    0.82
    quet
    0.82
    ju
    0.80
    awa
    0.80
    otation
    0.79
    vas
    0.77
    Activations Density 5.723%

    No Known Activations