© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 87
    Prev
    Next
    INDEX
    Explanations

    proper nouns of people, likely related to politics and criminal incidents

    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
    288
    +0.09
    0.5%
    326
    +0.09
    0.5%
    4
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.09
    0.07
    111
    +0.09
    0.06
    224
    +0.08
    0.06
    Negative Logits
     corrid
    -0.68
     millenn
    -0.65
     Paula
    -0.63
     Citiz
    -0.60
     reluct
    -0.57
     unden
    -0.57
    udeb
    -0.57
     signific
    -0.57
     welf
    -0.55
     });
    -0.54
    POSITIVE LOGITS
    enegger
    0.74
    yk
    0.70
    enson
    0.70
    lein
    0.69
     partName
    0.66
    arty
    0.64
    ci
    0.63
    auer
    0.63
    aux
    0.62
    ="#
    0.61
    Activations Density 0.420%

    No Known Activations