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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1646
    Prev
    Next
    INDEX
    Explanations

    disjointed pieces of text related to library operations

    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
    651
    +0.16
    0.5%
    38
    +0.14
    0.5%
    314
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.16
    0.05
    48
    +0.14
    0.04
    232
    +0.12
    0.03
    Negative Logits
    udic
    -1.17
    erer
    -1.10
    othal
    -0.95
    ERC
    -0.93
    paces
    -0.90
    ucer
    -0.90
    zzle
    -0.89
    uve
    -0.89
    aults
    -0.87
     graz
    -0.86
    POSITIVE LOGITS
    ··
    1.64
    µ
    1.09
    âĢ¢âĢ¢
    1.05
     Joined
    1.01
    ¼
    1.00
    eenth
    0.95
    Ħ
    0.95
    ·
    0.95
    ¾
    0.91
    ilities
    0.89
    Activations Density 0.506%

    No Known Activations