© 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. 9045
    Prev
    Next
    INDEX
    Explanations

    exclamatory sentences indicating excitement or urgency

    oai_token-act-pair · gpt-3.5-turbo

    exclamatory punctuation marks or emphatic statements

    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
    yrus
    -0.84
    undai
    -0.82
    idon
    -0.80
    mos
    -0.76
     manif
    -0.71
    jected
    -0.71
    itated
    -0.68
     footing
    -0.66
     plur
    -0.66
     restraints
    -0.65
    POSITIVE LOGITS
    @#&
    1.40
    #$
    1.00
     :)
    1.00
     :-)
    0.99
     Whilst
    0.96
    ?!
    0.92
     Additionally
    0.90
     ;)
    0.89
     Luckily
    0.88
     Also
    0.88
    Activations Density 0.070%

    No Known Activations