© 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. 10-RES-JB
    11. 2668
    Prev
    Next
    INDEX
    Explanations

    phrases related to thinking, analyzing, or considering various concepts or ideas

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing difficulty in recalling or thinking of examples or movements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yon
    -0.77
     pione
    -0.76
    encia
    -0.73
    yip
    -0.72
    shift
    -0.66
     teasp
    -0.66
    hift
    -0.66
    ju
    -0.65
    nels
    -0.65
    ysc
    -0.63
    POSITIVE LOGITS
     anyone
    1.02
     instances
    1.01
     anywhere
    1.01
     examples
    0.97
     anybody
    0.95
     precedent
    0.94
     instance
    0.94
     any
    0.91
     anything
    0.90
     comparable
    0.90
    Activations Density 0.184%

    No Known Activations