© 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. 21758
    Prev
    Next
    INDEX
    Explanations

    phrases related to physical characteristics and body parts

    oai_token-act-pair · gpt-3.5-turbo

    repetitive conjunctions and phrases emphasizing additions or similarities

    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
    igmatic
    -0.88
    Contents
    -0.83
    abul
    -0.79
    successfully
    -0.75
    ensable
    -0.74
    allel
    -0.73
    conduct
    -0.72
    escription
    -0.70
    iless
    -0.70
    cffffcc
    -0.68
    POSITIVE LOGITS
     blah
    1.51
     stuff
    1.31
     shit
    1.10
     thats
    1.07
     crap
    1.04
     maybe
    1.04
     THEN
    1.00
     secondly
    0.97
     everything
    0.96
     therefore
    0.92
    Activations Density 0.335%

    No Known Activations