© 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. 45829
    Prev
    Next
    INDEX
    Explanations

    formal language or legal terms

    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
    +0.17
    0.7%
    99
    +0.13
    0.6%
    146
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    447
    +0.17
    0.45
    138
    +0.13
    0.43
    64
    +0.11
    -0.21
    Negative Logits
    obyl
    -0.98
    acea
    -0.94
    atural
    -0.85
    ichick
    -0.84
    é¾įå
    -0.79
    anes
    -0.77
    ained
    -0.77
    aughlin
    -0.75
    olicited
    -0.75
    é¾
    -0.73
    POSITIVE LOGITS
    reth
    0.97
    rics
    0.83
    rers
    0.83
    ãĥ£
    0.81
    ive
    0.72
    rer
    0.71
    ressive
    0.70
    hered
    0.69
    ttes
    0.68
    cery
    0.68
    Activations Density 11.778%

    No Known Activations