© 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
    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. 1406
    Prev
    Next
    INDEX
    Explanations

    phrases related to various topics such as technology, sports, politics, and entertainment

    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
    320
    +0.17
    0.6%
    393
    +0.16
    0.5%
    513
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    582
    +0.17
    0.06
    598
    +0.16
    0.06
    168
    +0.14
    0.07
    Negative Logits
    him
    -1.08
    bard
    -1.04
    TM
    -1.01
    anamo
    -0.98
    kat
    -0.93
    von
    -0.92
    JJ
    -0.92
    mma
    -0.92
    Tan
    -0.90
    pas
    -0.89
    POSITIVE LOGITS
    etheless
    1.49
     nonetheless
    1.26
    »Ĵ
    1.03
     nevertheless
    1.03
     persisted
    1.03
     persists
    1.01
     beware
    0.98
     acknow
    0.95
     construed
    0.94
     conclud
    0.91
    Activations Density 1.206%

    No Known Activations