© 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. 19988
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "hammer" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "hammer" 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
    venture
    -0.75
     Citiz
    -0.74
    ventures
    -0.74
     nm
    -0.72
    Vill
    -0.72
     Surveillance
    -0.70
    iances
    -0.68
     VIP
    -0.67
    eries
    -0.67
    atri
    -0.66
    POSITIVE LOGITS
     hammer
    3.95
     Hammer
    1.94
    hammer
    1.93
     hamm
    1.88
     hammered
    1.68
     axe
    1.53
     spear
    1.38
     wrench
    1.36
     pounding
    1.27
     shovel
    1.21
    Activations Density 0.008%

    No Known Activations