© 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. 0-RES-JB
    11. 6797
    Prev
    Next
    INDEX
    Explanations

    comparisons of weight or brightness, particularly highlighting lighter options

    oai_token-act-pair · gpt-3.5-turbo

    references to "lighter" and "heavier" objects or qualities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Analytics
    -0.70
     Instit
    -0.69
     Coun
    -0.67
    abad
    -0.60
    Corp
    -0.58
     Scient
    -0.57
    plex
    -0.57
     Fol
    -0.57
    iev
    -0.57
    EMA
    -0.56
    POSITIVE LOGITS
     lighter
    3.81
     heavier
    2.40
     darker
    1.83
     thinner
    1.79
     brighter
    1.64
     thicker
    1.58
     softer
    1.54
    ighter
    1.51
     heaviest
    1.50
     light
    1.49
    Activations Density 0.014%

    No Known Activations