INDEX

Explanations

words that express positive sentiment or evaluation, or reference legal proceedings/outcomes

oai_token-act-pair · gemini-2.0-flash

positive sentiment

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zheim

-0.67

("]");

-0.61

-0.60

Thankyou

-0.57

 apologised

-0.57

addPreferredGap

-0.56

behaved

-0.56

endregion

-0.54

awtextra

-0.52

LikeLike

-0.51

POSITIVE LOGITS

<bos>

0.87

 occhiali

0.69

 '\\;'

0.69

 vestiti

0.59

 précisément

0.58

 färg

0.58

 hår

0.57

 geladeira

0.56

 الحره

0.56

 nigga

0.55

Activations Density 1.858%

words that express positive sentiment or evaluation, or reference legal proceedings/outcomes

positive sentiment

No Comments

No Known Activations