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 E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 13913
    Prev
    Next
    INDEX
    Explanations

    descriptions related to news articles and events

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    640
    +0.29
    2.1%
    480
    +0.10
    0.7%
    167
    +0.06
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.29
    0.39
    64
    +0.10
    -0.10
    94
    +0.06
    0.20
    Negative Logits
    boro
    -0.55
    ython
    -0.49
    ardless
    -0.47
    yton
    -0.46
    cius
    -0.46
     Reincarnated
    -0.44
     blinded
    -0.44
     endeavour
    -0.43
     placebo
    -0.43
    utic
    -0.43
    POSITIVE LOGITS
    ONSORED
    0.72
    ³³³³
    0.67
    ³³³
    0.65
    ³³³³³³³³³³³³³³³³
    0.63
    However
    0.61
    ³³³³³³³³
    0.58
    PHOTOS
    0.57
    Likewise
    0.57
    Nevertheless
    0.56
    VERTISEMENT
    0.56
    Activations Density 1.639%

    No Known Activations