Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 2-RES-JB
    6. 13
    Prev
    Next
    INDEX
    Explanations

    words related to physical tools or actions involving force

    oai_token-act-pair · gpt-3.5-turbo

    terms related to tools and mechanical actions that disrupt or interfere

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    words related to tools or actions that imply alteration or repair, particularly focusing on the root 'wrench'

    oai_token-act-pair · gpt-4-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cation
    -0.90
    itative
    -0.80
    vertis
    -0.78
    cious
    -0.74
    ité
    -0.73
    ombat
    -0.73
    phis
    -0.72
    veland
    -0.71
    anyon
    -0.71
    yrics
    -0.71
    POSITIVE LOGITS
     wrench
    1.17
     Schr
    1.12
     bitters
    0.81
     levers
    0.71
     fraught
    0.71
     pinch
    0.71
     wedge
    0.70
     scissors
    0.70
    driver
    0.68
     Sawyer
    0.66
    Activations Density 0.028%

    No Known Activations