© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 6967
    Prev
    Next
    INDEX
    Explanations

    references to individuals with specialized roles, particularly those related to law enforcement or security agencies

    oai_token-act-pair · gpt-3.5-turbo

    references to special agents and their roles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •Special

    No Comments

    Negative Logits
    yard
    -0.66
    enburg
    -0.66
    hari
    -0.62
    ylon
    -0.62
    shown
    -0.62
    estamp
    -0.60
    lis
    -0.60
    hurst
    -0.60
     Unlimited
    -0.59
    bane
    -0.59
    POSITIVE LOGITS
    ties
    1.03
    ised
    0.96
    ty
    0.95
     ops
    0.91
    izations
    0.90
     envoy
    0.89
     counsel
    0.87
    isite
    0.86
     Forces
    0.84
     prosecutor
    0.84
    Activations Density 0.031%

    No Known Activations