INDEX

Explanations

emphasizing words that indicate certainty or emphasis

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Sod

-0.65

Sod

-0.60

 hende

-0.59

สาย

-0.59

 serem

-0.59

 charging

-0.58

jap

-0.56

 scolaires

-0.55

 shaking

-0.54

 Shown

-0.54

POSITIVE LOGITS

 have

0.97

 gave

0.85

 could

0.84

verläs

0.82

had

0.82

$")

0.82

Šaltiniai

0.81

avía

0.81

 must

0.79

 nahilalakip

0.78

Activations Density 0.264%

emphasizing words that indicate certainty or emphasis

No Comments

No Known Activations