© 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. 10-RES-JB
    11. 23328
    Prev
    Next
    INDEX
    Explanations

    instances of the word "always" accompanied by other context

    oai_token-act-pair · gpt-3.5-turbo

    expressions of consistency or reliability

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yet
    -0.74
    alf
    -0.71
    aye
    -0.71
    ahime
    -0.70
    ajor
    -0.70
    dm
    -0.67
    folios
    -0.66
    ultan
    -0.66
    unal
    -0.65
    åĤ
    -0.63
    POSITIVE LOGITS
     lurking
    0.85
    theless
    0.81
    entimes
    0.80
     dreamed
    0.80
     strive
    0.80
     seemed
    0.78
     striving
    0.77
     wondered
    0.77
     bothered
    0.75
     underestimated
    0.74
    Activations Density 0.055%

    No Known Activations