© 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 · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 14195
    Prev
    Next
    INDEX
    Explanations

    phrases with the word "straight" with some being related to physical actions or attitudes

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    629
    +0.14
    0.5%
    1328
    +0.13
    0.4%
    197
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    629
    +0.14
    0.03
    197
    +0.13
    0.03
    1328
    +0.12
    0.02
    Negative Logits
     Renewed
    -0.63
     Enlarged
    -0.62
     Wtf
    -0.58
     affor
    -0.56
     effectually
    -0.56
     Conno
    -0.56
     richiede
    -0.55
     yow
    -0.55
     Mmm
    -0.55
     Shrub
    -0.54
    POSITIVE LOGITS
     straight
    1.14
    straight
    1.14
    Straight
    1.07
     Straight
    1.00
     STRA
    0.76
     strait
    0.64
    STRA
    0.64
     straighten
    0.62
    直
    0.57
     straightening
    0.57
    Activations Density 0.070%

    No Known Activations