© 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. 5261
    Prev
    Next
    INDEX
    Explanations

    commands and technical code syntax

    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
    393
    +0.14
    0.5%
    664
    +0.13
    0.4%
    380
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    639
    +0.14
    0.09
    556
    +0.13
    0.08
    314
    +0.12
    -0.00
    Negative Logits
    obyl
    -1.03
     disruptive
    -0.99
    ologies
    -0.95
     Osc
    -0.87
    ozy
    -0.86
     seams
    -0.86
     rhythms
    -0.85
    pmwiki
    -0.83
     rhythm
    -0.80
     curric
    -0.80
    POSITIVE LOGITS
    alty
    1.12
    andise
    1.04
    iversary
    1.02
    hoff
    1.02
    ÙĴ
    0.96
    intosh
    0.92
     Lauder
    0.91
    craft
    0.87
    amate
    0.85
    ée
    0.84
    Activations Density 1.467%

    No Known Activations