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. 5-RES-JB
    6. 2253
    Prev
    Next
    INDEX
    Explanations

    verbs indicating taking action or making decisions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "act" in various contexts related to taking action

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    paralle
    -0.72
    fer
    -0.71
    ciating
    -0.66
     Flavoring
    -0.66
     Begin
    -0.66
    hander
    -0.65
    lake
    -0.64
    clad
    -0.64
    xual
    -0.63
    gran
    -0.63
    POSITIVE LOGITS
    uate
    1.17
    uary
    1.14
    uated
    1.09
    inic
    1.04
    uating
    0.93
    ivism
    0.90
    ional
    0.87
    uations
    0.86
    uates
    0.86
    ivity
    0.85
    Activations Density 0.017%

    No Known Activations