© 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. 4-RES-JB
    11. 9686
    Prev
    Next
    INDEX
    Explanations

    references to sports positions, particularly the starting position

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "starter" in various contexts, particularly related to sports or cooking

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    erry
    -0.89
    oral
    -0.86
    illion
    -0.80
    fax
    -0.80
    orb
    -0.79
    romeda
    -0.78
    andra
    -0.77
    metic
    -0.74
    rams
    -0.70
    ores
    -0.69
    POSITIVE LOGITS
     starter
    1.12
     Starter
    0.84
    starter
    0.83
     pitcher
    0.78
     Rubin
    0.73
     starters
    0.73
    TAIN
    0.72
    FIELD
    0.69
    IOR
    0.69
     kit
    0.67
    Activations Density 0.008%

    No Known Activations