INDEX

Explanations

occurrences of the word "count" in various 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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Glaser

-0.55

🏻

-0.49

acağı

-0.47

ized

-0.47

 appena

-0.46

Dol

-0.46

>>>>>

-0.44

 esclusivamente

-0.44

 adequately

-0.44

Particular

-0.44

POSITIVE LOGITS

count

1.20

 count

1.17

Count

1.13

setCount

1.09

 Count

1.06

 counts

1.03

 Counts

0.96

COUNT

0.96

 COUNT

0.94

 totalCount

0.92

Activations Density 0.021%

occurrences of the word "count" in various contexts

No Comments

No Known Activations