© 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 Local
    9. 6-RES_SCL-AJT
    11. 5003
    Prev
    Next
    INDEX
    Explanations

    phrases related to decks, cards, and board games

    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
    447
    +0.16
    0.7%
    420
    +0.14
    0.6%
    167
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    526
    +0.16
    -0.06
    343
    +0.14
    0.07
    325
    +0.14
    0.07
    Negative Logits
    çīĪ
    -0.95
    heric
    -0.85
    phrine
    -0.76
    ipient
    -0.72
    ternity
    -0.71
    Downloadha
    -0.71
    interstitial
    -0.68
    ¥ŀ
    -0.67
     plur
    -0.66
    pheus
    -0.66
    POSITIVE LOGITS
    lists
    1.26
    yard
    1.09
    igans
    1.06
    wright
    1.06
    list
    1.02
    mates
    1.00
    igan
    0.99
    builder
    0.98
    hands
    0.95
    nut
    0.95
    Activations Density 7.838%

    No Known Activations