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. 13358
    Prev
    Next
    INDEX
    Explanations

    phrases related to programming tools or software development concepts

    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
    +0.30
    1.3%
    679
    +0.18
    0.8%
    420
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    60
    +0.30
    0.06
    653
    +0.18
    0.06
    679
    +0.14
    0.12
    Negative Logits
     Sons
    -0.91
     Reincarn
    -0.79
    oul
    -0.70
     nep
    -0.66
     adolesc
    -0.66
     Organic
    -0.66
     sons
    -0.66
     coli
    -0.66
     Pats
    -0.65
     Pictures
    -0.65
    POSITIVE LOGITS
    tips
    1.64
    kit
    1.52
    tip
    1.25
    sonian
    1.11
    belt
    1.09
     tools
    1.01
    chain
    1.01
     tool
    0.98
    box
    0.98
    chains
    0.97
    Activations Density 5.711%

    No Known Activations