© 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. 45832
    Prev
    Next
    INDEX
    Explanations

    references to weapons and weapon-related concepts

    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
    +0.28
    1.2%
    288
    +0.19
    0.8%
    128
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    128
    +0.28
    0.10
    526
    +0.19
    -0.06
    370
    +0.11
    0.06
    Negative Logits
    gres
    -0.87
     coli
    -0.81
    noon
    -0.78
    lus
    -0.78
    cess
    -0.76
    forth
    -0.76
    apest
    -0.74
     Strawberry
    -0.74
     Niet
    -0.73
    atta
    -0.72
    POSITIVE LOGITS
    powder
    1.02
     salute
    0.91
     arsenal
    0.88
     weapons
    0.85
     caches
    0.85
     wielded
    0.85
     weapon
    0.84
     racks
    0.83
     manufacturer
    0.82
    kit
    0.79
    Activations Density 8.259%

    No Known Activations