© 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. 0-RES-JB
    11. 22325
    Prev
    Next
    INDEX
    Explanations

    numbers that are likely to be Twitter handles or usernames

    oai_token-act-pair · gpt-3.5-turbo

    specific usernames or identifiers related to social media or content creators

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Rivera
    -0.85
     Mercenary
    -0.71
     Squad
    -0.71
     Fury
    -0.67
    REDACTED
    -0.67
     Kush
    -0.65
    ibaba
    -0.64
    Factor
    -0.63
     Dynasty
    -0.62
    KI
    -0.62
    POSITIVE LOGITS
    ļéĨĴ
    0.72
    lyak
    0.70
    abies
    0.68
    unte
    0.66
    byn
    0.65
     differed
    0.65
    thodox
    0.64
    aste
    0.64
    itutional
    0.63
    rose
    0.63
    Activations Density 0.001%

    No Known Activations