© 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. 11-RES-JB
    11. 16355
    Prev
    Next
    INDEX
    Explanations

    measurements or dimensions represented in feet

    oai_token-act-pair · gpt-3.5-turbo

    measurements and dimensions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     physic
    -0.86
     physi
    -0.72
     controls
    -0.71
     Proc
    -0.70
     affili
    -0.70
     prob
    -0.67
     control
    -0.66
     context
    -0.66
     observer
    -0.66
     ent
    -0.66
    POSITIVE LOGITS
    tall
    1.59
    plus
    1.48
    long
    1.48
    per
    1.47
    high
    1.38
    deep
    1.36
    sized
    1.36
    million
    1.32
    interstitial
    1.31
    gross
    1.30
    Activations Density 0.047%

    No Known Activations