© 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. 3-RES-JB
    11. 10771
    Prev
    Next
    INDEX
    Explanations

    instances of the word "hitter" with activations indicating different levels of relevance

    oai_token-act-pair · gpt-3.5-turbo

    references to baseball players and their performance, particularly hitters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iment
    -0.99
    ¥µ
    -0.88
    rex
    -0.81
    chin
    -0.79
     Imperium
    -0.72
    ulla
    -0.71
    iments
    -0.70
    ional
    -0.70
    qual
    -0.69
    flag
    -0.69
    POSITIVE LOGITS
     batted
    1.12
     batters
    1.07
     hitters
    1.07
     batting
    1.05
     hitter
    1.04
     clubhouse
    0.94
     outfielder
    0.94
     pitchers
    0.89
     pitcher
    0.87
    BIP
    0.87
    Activations Density 0.027%

    No Known Activations