© 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
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 3890
    Prev
    Next
    INDEX
    Explanations

    tournament-related phrases and numerical values

    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
    687
    +0.08
    0.6%
    285
    +0.07
    0.5%
    398
    +0.06
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    268
    +0.08
    0.01
    132
    +0.07
    0.01
    391
    +0.06
    0.01
    Negative Logits
    bre
    -0.47
     Jonas
    -0.45
    roy
    -0.43
    mong
    -0.43
     Dough
    -0.42
     404
    -0.42
     Pulitzer
    -0.42
     Stockholm
    -0.42
    tur
    -0.40
    rams
    -0.40
    POSITIVE LOGITS
    eatures
    0.45
     encount
    0.45
    ÙĴ
    0.43
    Component
    0.43
    ighth
    0.42
    ioxide
    0.42
     corrid
    0.41
    assian
    0.41
     Besides
    0.40
     "$:/
    0.40
    Activations Density 0.007%

    No Known Activations