© 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. 14070
    Prev
    Next
    INDEX
    Explanations

    phrases indicating uncertainty or necessity in relation to statements made

    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
    151
    +0.16
    0.5%
    132
    +0.14
    0.4%
    208
    +0.14
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    151
    +0.16
    0.12
    307
    +0.14
    0.09
    718
    +0.14
    0.10
    Negative Logits
    YES
    -1.01
    abre
    -1.00
     Belt
    -0.96
    lain
    -0.96
    Pages
    -0.94
    dogs
    -0.91
     Repe
    -0.91
     MAY
    -0.91
    Neigh
    -0.90
     Bomber
    -0.90
    POSITIVE LOGITS
     bothered
    1.34
     anymore
    1.33
    icably
    1.23
     discern
    1.22
     bother
    1.20
     inclined
    1.19
    icable
    1.18
     detract
    1.14
     harmed
    1.12
     distinguish
    1.12
    Activations Density 1.958%

    No Known Activations