© 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. 3-RES-JB
    11. 1717
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a person speaking or giving a speech

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "spoke" indicating reported speech or dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    assi
    -0.71
    gradation
    -0.68
    mental
    -0.66
    \/\/
    -0.66
    vernment
    -0.66
    =~
    -0.65
    inner
    -0.65
    engeance
    -0.65
    rypt
    -0.65
    ======
    -0.65
    POSITIVE LOGITS
     aloud
    0.93
    ophone
    0.85
     artic
    0.83
     volumes
    0.82
     voices
    0.80
     spoken
    0.78
     fluent
    0.77
     Cosponsors
    0.77
     Mandarin
    0.76
    voice
    0.76
    Activations Density 0.024%

    No Known Activations