© 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. 106
    Prev
    Next
    INDEX
    Explanations

    phrases related to scientific and technical information

    oai_token-act-pair · gpt-3.5-turbo

    phrases or concepts related to time, particularly compound words or phrases containing "time" (e.g. "time-varying", "time-consuming", "time-dependent").

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot
    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
    136
    +0.15
    0.5%
    340
    +0.12
    0.4%
    568
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    546
    +0.15
    0.04
    678
    +0.12
    0.04
    280
    +0.12
    0.04
    Negative Logits
    Wan
    -1.14
    fml
    -1.06
    æĦ
    -1.03
     Flavoring
    -1.03
    Filename
    -1.03
    Chicken
    -1.02
    uay
    -0.97
    Da
    -0.96
    çĶŁ
    -0.95
    Bi
    -0.95
    POSITIVE LOGITS
    ourses
    1.15
    anwhile
    1.03
     commute
    1.00
    consuming
    0.98
     incarcer
    0.96
     trave
    0.96
    boarding
    0.95
     adjourn
    0.95
    uploads
    0.91
     intervals
    0.90
    Activations Density 1.568%

    No Known Activations