© 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. 23670
    Prev
    Next
    INDEX
    Explanations

    phrases indicating contrast or disagreement

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "No" followed by varying degrees of emphasis regarding certainty or agreement

    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
    iership
    -0.77
    ials
    -0.71
    ãĤ¼ãĤ¦ãĤ¹
    -0.70
    RAFT
    -0.66
    lycer
    -0.66
    rex
    -0.66
    romy
    -0.65
    endish
    -0.64
    assies
    -0.64
    aven
    -0.62
    POSITIVE LOGITS
    xious
    1.05
     longer
    0.97
     matter
    0.96
    etheless
    0.93
     kidding
    0.89
     doubt
    0.88
    zzle
    0.85
     indication
    0.84
     clue
    0.83
    terday
    0.81
    Activations Density 0.063%

    No Known Activations