© 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. 2-RES-JB
    11. 647
    Prev
    Next
    INDEX
    Explanations

    the word "acting" regardless of the context in which it appears

    oai_token-act-pair · gpt-4-turbo

    mentions of the action of acting in the context of performing roles in various settings

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the act of performing in theater or film

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    xual
    -0.84
    joy
    -0.73
    foot
    -0.71
    nda
    -0.68
    origin
    -0.66
    lake
    -0.66
    tch
    -0.65
    ilings
    -0.64
    issue
    -0.64
    front
    -0.63
    POSITIVE LOGITS
    uate
    0.84
    uary
    0.82
    ional
    0.81
    uated
    0.78
     acting
    0.74
    encies
    0.74
    uations
    0.73
    iate
    0.71
    entials
    0.67
    units
    0.67
    Activations Density 0.008%

    No Known Activations