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. 4260
    Prev
    Next
    INDEX
    Explanations

    updates in articles

    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
    202
    +0.10
    0.5%
    559
    +0.09
    0.5%
    219
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.10
    0.23
    393
    +0.09
    0.25
    362
    +0.09
    0.10
    Negative Logits
    pmwiki
    -0.73
    æ©
    -0.59
    advant
    -0.58
    gling
    -0.55
    ENG
    -0.55
    WOOD
    -0.55
    ãĤ¼ãĤ¦ãĤ¹
    -0.53
    zees
    -0.53
    ymm
    -0.53
    SPONSORED
    -0.52
    POSITIVE LOGITS
     updates
    0.77
    dates
    0.68
     updated
    0.68
     chron
    0.67
     reflect
    0.64
    update
    0.63
     update
    0.62
     comments
    0.62
    patch
    0.62
     batches
    0.61
    Activations Density 10.376%

    No Known Activations