INDEX

Explanations

phrases emphasizing the word "so" and its variations in connection with context or consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

ondissement

-0.53

illah

-0.52

τεύ

-0.47

iastes

-0.44

zegorz

-0.43

PDOException

-0.43

TypedDataSet

-0.42

országban

-0.41

enaar

-0.41

 Aner

-0.41

POSITIVE LOGITS

did

0.95

 does

0.91

AndEndTag

0.85

 saites

0.81

Did

0.77

aarrggbb

0.75

did

0.74

 Does

0.73

Did

0.71

do

0.70

Activations Density 0.180%

phrases emphasizing the word "so" and its variations in connection with context or consequences

No Comments

No Known Activations