INDEX

Explanations

phrases related to official statements and comments

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

Follows "I" or quotation marks

np_acts-logits-general · gemini-2.0-flash

comment refusing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

 aans

-0.43

doubt

-0.41

 cromado

-0.39

calça

-0.39

 Doubt

-0.39

Doubt

-0.38

 חיצוניים

-0.37

rawtypes

-0.37

 proposte

-0.37

worse

-0.37

POSITIVE LOGITS

 declined

0.75

 comment

0.72

 commenting

0.60

 details

0.59

 declines

0.55

 comments

0.55

 decline

0.54

 Decline

0.52

 Comment

0.52

 refused

0.51

Activations Density 0.165%

phrases related to official statements and comments

Follows "I" or quotation marks

comment refusing

No Comments

No Known Activations