© 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. 8-RES-JB
    11. 8973
    Prev
    Next
    INDEX
    Explanations

    commas used as a pause in sentences

    oai_token-act-pair · gpt-3.5-turbo

    phrases that introduce or qualify information, particularly emphasizing contrasts or clarifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cott
    -0.61
    "]=>
    -0.59
    izon
    -0.54
    minecraft
    -0.53
    ORPG
    -0.52
    uge
    -0.52
    num
    -0.52
    da
    -0.51
    ift
    -0.51
    ield
    -0.51
    POSITIVE LOGITS
     albeit
    0.96
     however
    0.91
     moreover
    0.89
     alas
    0.83
     somew
    0.80
     therefore
    0.74
     anecd
    0.71
    jri
    0.71
     though
    0.70
     incidentally
    0.70
    Activations Density 0.229%

    No Known Activations