© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 3858
    Prev
    Next
    INDEX
    Explanations

    information related to police operations, arrests, SWAT teams, and military activities, including tools, events, and outcomes

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    330
    +0.12
    0.5%
    193
    +0.12
    0.4%
    34
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    537
    +0.12
    0.11
    762
    +0.12
    0.14
    330
    +0.11
    0.09
    Negative Logits
    ãĥ´ãĤ¡
    -0.88
     unification
    -0.85
     grievance
    -0.84
    ¥µ
    -0.83
    ãĥ´
    -0.80
    title
    -0.79
    ãĤ¨ãĥ«
    -0.78
     repentance
    -0.78
     Childhood
    -0.77
    âĸ¬
    -0.76
    POSITIVE LOGITS
     stationed
    1.54
     deployed
    1.32
     patrolling
    1.30
     dispatched
    1.17
     positioned
    1.16
     swoop
    1.13
     tasked
    1.10
     hired
    1.06
     trained
    1.06
     excav
    1.06
    Activations Density 4.424%

    No Known Activations