© 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. 1-RES-JB
    11. 2730
    Prev
    Next
    INDEX
    Explanations

    phrases related to responses or reactions to various stimuli

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a reaction or answer to a situation or event

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cutting
    -0.73
    rafted
    -0.72
    hemat
    -0.69
    ju
    -0.68
    rust
    -0.66
    ramer
    -0.63
     knots
    -0.62
    ffe
    -0.62
     sliced
    -0.61
    rome
    -0.60
    POSITIVE LOGITS
     response
    1.25
     responses
    1.11
    response
    1.09
    reply
    0.98
    DragonMagazine
    0.97
     respond
    0.97
     Response
    0.96
     reply
    0.93
     responders
    0.90
     reaction
    0.88
    Activations Density 0.020%

    No Known Activations