INDEX

Explanations

phrases indicating annoyance or dissatisfaction

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

 only

-0.16

 exact

-0.16

 absolutely

-0.16

 actually

-0.16

 absolute

-0.16

 direct

-0.16

 full

-0.15

 almost

-0.15

 both

-0.15

 exactly

-0.14

POSITIVE LOGITS

 harmless

0.25

 mere

0.23

isol

0.23

mere

0.21

 ordinary

0.20

 normal

0.20

 isolated

0.20

 passing

0.19

 inconvenience

0.19

æŃ£å¸¸

0.19

Activations Density 0.094%

phrases indicating annoyance or dissatisfaction

No Comments

No Known Activations