© 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. 45959
    Prev
    Next
    INDEX
    Explanations

    phrases related to government and political theories

    oai_token-act-pair · gpt-3.5-turbo

    words that begin with prefixes related to forgiving or forgetting, including forms of the root "forgive" or similar constructs

    oai_token-act-pair · gpt-3.5-turboTriggered by @danbraun
    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
    526
    +0.08
    0.6%
    239
    +0.08
    0.6%
    289
    +0.07
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.08
    0.22
    352
    +0.08
    0.13
    289
    +0.07
    0.00
    Negative Logits
    rons
    -0.47
    ries
    -0.43
    rice
    -0.41
     clerks
    -0.40
     instincts
    -0.39
     Weiner
    -0.38
     ruling
    -0.38
     entrances
    -0.38
     directions
    -0.37
     result
    -0.37
    POSITIVE LOGITS
    ets
    0.61
    ibility
    0.58
    oul
    0.55
    igning
    0.54
    eme
    0.51
    DonaldTrump
    0.50
    icker
    0.50
    igned
    0.50
    pler
    0.49
    eling
    0.49
    Activations Density 7.731%

    No Known Activations