INDEX

Explanations

occurrences of the word "word" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Clifton

-0.55

 Helf

-0.52

Hau

-0.49

IANI

-0.49

uley

-0.48

 Gauthier

-0.48

ilaire

-0.47

 mistrust

-0.47

$:$

-0.46

atelli

-0.46

POSITIVE LOGITS

Word

1.13

 word

1.13

 Word

1.08

 WORD

1.08

word

1.01

 words

0.98

Words

0.96

 WORDS

0.96

WORD

0.96

 Words

0.93

Activations Density 0.125%

occurrences of the word "word" and its variations in different contexts

No Comments

No Known Activations