© 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. 19947
    Prev
    Next
    INDEX
    Explanations

    words related to difficulties or challenges

    oai_token-act-pair · gpt-3.5-turbo

    references to various real-world challenges or difficulties

    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
    tein
    -0.82
    ride
    -0.73
    wark
    -0.71
     Vide
    -0.69
     salute
    -0.68
    pire
    -0.68
    CG
    -0.65
    tyard
    -0.65
     homage
    -0.65
    ifter
    -0.64
    POSITIVE LOGITS
     problems
    3.42
     Problems
    2.72
     difficulties
    2.49
     issues
    2.37
     troubles
    2.23
     problem
    2.20
     woes
    2.20
    problem
    2.01
     Problem
    1.96
     deficiencies
    1.93
    Activations Density 0.030%

    No Known Activations