© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 8376
    Prev
    Next
    INDEX
    Explanations

    detailed and comprehensive descriptions or analyses in a document

    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
    195
    +0.22
    0.8%
    99
    +0.14
    0.5%
    302
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.22
    0.04
    455
    +0.14
    0.08
    302
    +0.13
    0.07
    Negative Logits
    lua
    -0.98
    warm
    -0.97
    leader
    -0.90
    thur
    -0.89
    haw
    -0.88
    enos
    -0.88
    opers
    -0.88
    corn
    -0.87
    hops
    -0.87
    ques
    -0.86
    POSITIVE LOGITS
    bred
    1.19
    mble
    1.11
    rehensive
    1.10
     detail
    1.09
    sonian
    1.01
     coverage
    1.00
     overhaul
    0.97
    lly
    0.97
     detailing
    0.97
     documentation
    0.96
    Activations Density 1.185%

    No Known Activations