© 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. 17297
    Prev
    Next
    INDEX
    Explanations

    phrases related to establishing and influencing the direction or mood

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to setting standards or precedents

    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
    leness
    -0.70
    ividual
    -0.69
    ¬¼
    -0.67
     outweigh
    -0.67
    vernment
    -0.62
    ugg
    -0.60
     outwe
    -0.59
     derive
    -0.58
    jug
    -0.58
    enza
    -0.58
    POSITIVE LOGITS
     precedent
    1.17
    tle
    1.05
     tone
    1.02
     sights
    1.00
    flame
    0.97
     precedence
    0.97
     preced
    0.97
     abl
    0.92
     groundwork
    0.90
     pace
    0.86
    Activations Density 0.062%

    No Known Activations