© 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. 21449
    Prev
    Next
    INDEX
    Explanations

    specific mentions of locations or entities in "what's going on" contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to specific locations or events occurring in various places

    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
    Not in Any Lists

    No Comments

    Negative Logits
    hler
    -0.73
    iculty
    -0.72
    username
    -0.72
    ensity
    -0.69
    ool
    -0.68
    casters
    -0.68
    angered
    -0.67
    rored
    -0.66
    ebus
    -0.66
    ools
    -0.65
    POSITIVE LOGITS
     regards
    1.16
     relation
    1.12
     terms
    0.98
     regard
    0.96
     milliseconds
    0.80
     practice
    0.79
     hindsight
    0.78
    terms
    0.76
     practise
    0.72
     retrospect
    0.72
    Activations Density 0.225%

    No Known Activations