INDEX

Explanations

phrases that express judgment, evaluation, or consideration

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

New Auto-Interp

Configuration

Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post

Features

65,536

Data Type

float32

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

gated

Context Size

1,024

Dataset

Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Gallagher

-0.15

jak

-0.15

 tentative

-0.15

 snapping

-0.14

atat

-0.14

imb

-0.14

paged

-0.13

 permission

-0.13

If

-0.13

 Basket

-0.13

POSITIVE LOGITS

fit

0.43

-fit

0.29

 necessary

0.29

fit

0.28

 appropriate

0.28

 Necessary

0.26

Fit

0.26

Fit

0.25

necessary

0.25

appropriate

0.25

Activations Density 0.013%

phrases that express judgment, evaluation, or consideration

No Comments

No Known Activations