© 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. 22253
    Prev
    Next
    INDEX
    Explanations

    text related to physical deformation or structural changes

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "formation" and variations of the word "pairing."

    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
     MISS
    -0.69
     DPR
    -0.69
    govtrack
    -0.68
    erala
    -0.67
    umped
    -0.67
    ãĤ¶
    -0.65
     SPACE
    -0.65
     Mehran
    -0.64
     JUL
    -0.64
    oshenko
    -0.63
    POSITIVE LOGITS
    formation
    1.94
     pairing
    1.39
     Fit
    1.28
     domest
    1.19
     Piper
    1.14
    Fit
    1.09
     compass
    1.04
    medi
    0.99
    uo
    0.98
     mentions
    0.95
    Activations Density 0.053%

    No Known Activations