© 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 Local
    5. 6-RES_SCL-AJT
    6. 4030
    Prev
    Next
    INDEX
    Explanations

    descriptions of photos in an article

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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
    398
    +0.08
    0.5%
    683
    +0.08
    0.5%
    373
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.08
    -0.14
    398
    +0.08
    0.06
    739
    +0.08
    0.03
    Negative Logits
    owe
    -0.55
    manship
    -0.54
    truth
    -0.52
    Ħ¢
    -0.52
    ©¶æ¥µ
    -0.49
    shaw
    -0.48
    lli
    -0.47
    omsky
    -0.47
     rag
    -0.45
    swing
    -0.45
    POSITIVE LOGITS
     Uriel
    0.57
     IMAGES
    0.54
     Thumbnails
    0.50
    ãĥ´ãĤ¡
    0.49
     restrooms
    0.46
     VIDEOS
    0.46
     thumbnail
    0.46
    Interstitial
    0.45
     Personnel
    0.45
     portrait
    0.45
    Activations Density 5.767%

    No Known Activations