© 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. 14192
    Prev
    Next
    INDEX
    Explanations

    Lisp code structures such as nested parentheses and specific keywords

    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
    +0.26
    0.9%
    160
    +0.20
    0.7%
    314
    +0.16
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.26
    0.10
    314
    +0.20
    0.03
    640
    +0.16
    0.08
    Negative Logits
    heit
    -1.20
     Timbers
    -0.89
    umph
    -0.84
    inois
    -0.84
     hypers
    -0.81
    isodes
    -0.81
     Renew
    -0.78
     Crusade
    -0.77
     grap
    -0.76
    leep
    -0.76
    POSITIVE LOGITS
    ...)
    1.33
    âĢ¦)
    1.31
    ?)
    1.30
    *)
    1.29
    ,)
    1.22
    ++)
    1.20
    aka
    1.17
    Introduced
    1.11
    emphasis
    1.11
    ?),
    1.10
    Activations Density 0.784%

    No Known Activations