© 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. 12771
    Prev
    Next
    INDEX
    Explanations

    words related to data analysis and statistical procedures

    oai_token-act-pair · gpt-3.5-turbo

    terms related to quantitative data analysis and statistical methods

    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
     Philly
    -0.74
    Kids
    -0.69
     Deals
    -0.68
     daddy
    -0.68
     Chattanooga
    -0.68
    Yep
    -0.68
     cops
    -0.67
     Kids
    -0.67
     Parkway
    -0.67
     Daddy
    -0.66
    POSITIVE LOGITS
     uncertainties
    1.04
     analyses
    1.04
     hypotheses
    1.03
     comput
    1.02
     phylogen
    1.01
     âĪ¼
    1.01
     eluc
    1.00
     inferred
    0.99
    Methods
    0.99
     criterion
    0.99
    Activations Density 0.383%

    No Known Activations