© 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. 4-RES-JB
    11. 9077
    Prev
    Next
    INDEX
    Explanations

    words related to early time periods or occurrences

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "early" and similar variations indicating timeliness or the beginning of events

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

    No Comments

    Negative Logits
    retched
    -0.76
    uyomi
    -0.75
    atism
    -0.72
    ittee
    -0.70
    olulu
    -0.67
    ortium
    -0.66
    mbuds
    -0.66
    edIn
    -0.65
    shall
    -0.64
    opted
    -0.63
    POSITIVE LOGITS
     adop
    0.89
     stages
    0.83
     onset
    0.78
     drafts
    0.73
     phases
    0.73
    prototype
    0.71
     morning
    0.71
    -'
    0.70
    bird
    0.69
     mornings
    0.69
    Activations Density 0.038%

    No Known Activations