© 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. 0-RES-JB
    11. 1294
    Prev
    Next
    INDEX
    Explanations

    phrases related to disapproval or condemnation

    oai_token-act-pair · gpt-3.5-turbo

    terms related to fields and conditions associated with specific contexts or environments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    endo
    -0.70
    olution
    -0.65
    otten
    -0.63
    prison
    -0.63
    psy
    -0.63
     Pun
    -0.63
     prote
    -0.63
    lip
    -0.61
     rewritten
    -0.60
    uries
    -0.60
    POSITIVE LOGITS
    FIELD
    3.74
     depl
    1.30
     fide
    1.30
    edia
    1.29
    BACK
    1.15
    hari
    1.12
    INGTON
    1.00
    heric
    0.97
     BELOW
    0.90
    GROUND
    0.85
    Activations Density 0.060%

    No Known Activations