© 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. 3703
    Prev
    Next
    INDEX
    Explanations

    phrases related to the recruitment and transfer of football (soccer) players

    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
    393
    +0.34
    1.0%
    640
    +0.18
    0.5%
    552
    +0.16
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.34
    0.08
    589
    +0.18
    0.05
    755
    +0.16
    0.05
    Negative Logits
    CHAT
    -1.06
    Champ
    -1.05
    士
    -1.04
    ptives
    -0.94
    irection
    -0.93
    hots
    -0.92
    awei
    -0.92
    IPP
    -0.91
    soDeliveryDate
    -0.91
    TION
    -0.91
    POSITIVE LOGITS
     compensate
    1.50
     enhance
    1.34
     counteract
    1.28
     shore
    1.28
     strengthen
    1.25
     clarify
    1.24
     improve
    1.24
     bolster
    1.24
     alleviate
    1.23
     appease
    1.23
    Activations Density 0.735%

    No Known Activations