© 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 Local
    9. 6-RES_SCL-AJT
    11. 40062
    Prev
    Next
    INDEX
    Explanations

    Russian characters like "е" and "Ñĥ" that form part of proper nouns or names

    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
    +0.09
    0.5%
    503
    +0.09
    0.5%
    270
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    327
    +0.09
    0.13
    48
    +0.09
    0.12
    754
    +0.08
    0.12
    Negative Logits
    enna
    -0.71
    etts
    -0.69
    aneers
    -0.69
    ierrez
    -0.67
    ativity
    -0.65
    itches
    -0.63
    htaking
    -0.62
    ĸļ
    -0.62
    iggs
    -0.60
    rils
    -0.60
    POSITIVE LOGITS
    士
    0.88
    姫
    0.80
    ufact
    0.78
    ãĤ¨ãĥ«
    0.72
    åİ
    0.71
    æĹ
    0.66
    âĶĢâĶĢ
    0.66
    ises
    0.65
    vel
    0.65
    ULT
    0.64
    Activations Density 5.402%

    No Known Activations