© 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. 20022
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to sports teams, particularly those with the word "Stars" in them

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Stars" in various contexts

    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
     burglary
    -0.67
     kidney
    -0.65
    cu
    -0.63
    cro
    -0.63
    farm
    -0.63
     renal
    -0.62
     IE
    -0.62
    legal
    -0.61
     pizza
    -0.60
     breaching
    -0.60
    POSITIVE LOGITS
     Stars
    4.20
    Stars
    2.86
    stars
    2.28
     stars
    2.17
     STAR
    1.59
     Star
    1.52
    star
    1.52
     star
    1.38
     Starship
    1.37
    Star
    1.33
    Activations Density 0.013%

    No Known Activations