© 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. 10071
    Prev
    Next
    INDEX
    Explanations

    mentions of strong emotions of interest or enthusiasm

    oai_token-act-pair · gpt-3.5-turbo

    expressions of strong enthusiasm or dedication towards activities or interests

    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
    pta
    -0.77
    å¸
    -0.76
    annis
    -0.76
    eor
    -0.74
    avis
    -0.73
    ramid
    -0.66
     WATCHED
    -0.65
    reens
    -0.64
    ãĥĥãĥī
    -0.64
     helicop
    -0.63
    POSITIVE LOGITS
     passion
    1.05
    atical
    1.03
     passionately
    0.91
    iously
    0.86
     passionate
    0.84
     passions
    0.83
    uous
    0.83
    uality
    0.82
    ful
    0.80
    edIn
    0.80
    Activations Density 0.014%

    No Known Activations