© 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. 32979
    Prev
    Next
    INDEX
    Explanations

    specific phrases related to gaming strategies and actions

    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
    +0.11
    0.6%
    91
    +0.11
    0.5%
    88
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    193
    +0.11
    0.09
    760
    +0.11
    0.08
    570
    +0.10
    -0.00
    Negative Logits
     barrier
    -0.66
    çĽ
    -0.62
    åº
    -0.61
     Mul
    -0.60
    ãĥĥãĥĪ
    -0.59
     diffusion
    -0.59
    çIJ
    -0.59
     McGr
    -0.59
    apt
    -0.57
     Duc
    -0.57
    POSITIVE LOGITS
    !.
    0.83
    yah
    0.82
    !,
    0.76
    !'
    0.70
    qq
    0.68
    !:
    0.67
    !/
    0.67
    !",
    0.67
    FX
    0.66
    !),
    0.65
    Activations Density 8.281%

    No Known Activations