Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 13362
    Prev
    Next
    INDEX
    Explanations

    phrases related to individuals holding a strong position or belief, particularly in the context of politics or decision-making

    oai_token-act-pair · gpt-3.5-turbo

    verbs associated with maintaining or keeping something in a specific state or condition

    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
    346
    +0.13
    0.5%
    216
    +0.13
    0.5%
    183
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    183
    +0.13
    0.09
    566
    +0.13
    0.08
    582
    +0.12
    0.05
    Negative Logits
    ibel
    -0.82
    shire
    -0.81
    ————
    -0.81
    mph
    -0.78
     Daylight
    -0.77
    naire
    -0.77
    ologist
    -0.75
    endix
    -0.75
    gnu
    -0.75
    audi
    -0.74
    POSITIVE LOGITS
    hold
    1.31
    holding
    1.15
    holder
    1.15
    erness
    1.14
     sway
    1.14
    holders
    1.12
    held
    1.04
     hold
    1.03
     accountable
    0.98
    overs
    0.95
    Activations Density 7.582%

    No Known Activations