INDEX

Explanations

punctuation and structural markers used within sentences

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

crocheters

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 transfieras

-0.46

soort

-0.43

delwed

-0.42

 OMITBAD

-0.40

 Vergrößern

-0.40

 soort

-0.40

̈́

-0.40

beforeEach

-0.40

OrNil

-0.39

期刊论文

-0.39

POSITIVE LOGITS

ſelf

0.60

 itſelf

0.56

 himſelf

0.54

 pleaſure

0.52

support

0.51

 autorytatywna

0.51

Support

0.50

 Chriftian

0.49

 Majefty

0.48

 ſche

0.47

Activations Density 0.022%

punctuation and structural markers used within sentences

crocheters

No Comments

No Known Activations

punctuation and structural markers used within sentences

crocheters

No Comments

No Known Activations