© 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. 12736
    Prev
    Next
    INDEX
    Explanations

    mentions of specific organizations or groups and their positions or actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include commas, likely indicating lists or complex structures in sentences

    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
    animate
    -0.69
    DragonMagazine
    -0.63
    -,
    -0.58
    ãĥij
    -0.58
    worldly
    -0.57
     deed
    -0.57
    rats
    -0.56
    fw
    -0.56
     redundant
    -0.54
     ðŁĻĤ
    -0.54
    POSITIVE LOGITS
     disagrees
    1.13
     agrees
    1.07
     tells
    1.07
     says
    1.06
     told
    1.05
    said
    1.04
     argues
    1.04
     testified
    1.03
     believes
    1.03
     contends
    1.01
    Activations Density 0.138%

    No Known Activations