© 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
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 5315
    Prev
    Next
    INDEX
    Explanations

    phrases related to ground-based activities or concepts

    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
    +0.28
    1.0%
    667
    +0.14
    0.5%
    649
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.28
    0.06
    268
    +0.14
    0.03
    289
    +0.14
    0.02
    Negative Logits
    ansen
    -1.07
    azi
    -1.03
    icer
    -0.98
    anos
    -0.96
    itton
    -0.96
    aird
    -0.95
    volent
    -0.94
    aths
    -0.92
    ciation
    -0.90
    pload
    -0.90
    POSITIVE LOGITS
    hog
    1.58
    sheet
    1.19
    spring
    1.13
    nut
    1.12
    sheets
    1.10
    locked
    1.10
    notes
    1.09
    fare
    1.08
    nuts
    1.02
    cover
    0.98
    Activations Density 0.673%

    No Known Activations