INDEX

Explanations

comparative and superlative forms of adjectives

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

following "more"

np_acts-logits-general · gemini-2.0-flash

more followed by adjective

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 very

-0.59

muito

-0.56

 очень

-0.56

 velmi

-0.54

 muito

-0.54

 매우

-0.52

muy

-0.52

非常に

-0.52

 rất

-0.51

very

-0.50

POSITIVE LOGITS

bootstrapcdn

0.63

 likely

0.63

tagHelperRunner

0.60

 inclined

0.59

+#+#

0.57

IsMutable

0.57

 autorytatywna

0.56

WithIOException

0.56

הערות

0.56

Története

0.56

Activations Density 0.313%

comparative and superlative forms of adjectives

following "more"

more followed by adjective

No Known Activations