© 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. 33152
    Prev
    Next
    INDEX
    Explanations

    phrases related to power dynamics, politics, and societal issues

    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
    549
    +0.09
    0.5%
    283
    +0.09
    0.5%
    644
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.09
    0.22
    48
    +0.09
    0.12
    288
    +0.08
    0.21
    Negative Logits
    Ò
    -0.67
     Nun
    -0.61
     Hots
    -0.60
     Hague
    -0.59
    ··
    -0.56
    ĨĴ
    -0.56
     Lynd
    -0.56
    rent
    -0.53
     Webster
    -0.52
    ;;;;;;;;;;;;
    -0.52
    POSITIVE LOGITS
    cles
    1.03
    ulously
    0.97
    enza
    0.85
    uled
    0.83
    rics
    0.78
    cled
    0.77
    uling
    0.72
    uler
    0.72
    cling
    0.72
    ulous
    0.71
    Activations Density 6.565%

    No Known Activations