INDEX

Explanations

occurrences of the word "so."

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

so

-0.24

pane

-0.16

 sÃ¥

-0.16

neo

-0.16

ÑĭÑĪ

-0.15

panies

-0.15

noon

-0.15

roman

-0.15

plain

-0.15

mps

-0.15

POSITIVE LOGITS

-called

0.40

far

0.32

ething

0.28

iled

0.28

apy

0.28

 forth

0.26

zial

0.25

far

0.24

oner

0.24

ars

0.23

Activations Density 0.096%

occurrences of the word "so."

No Comments

No Known Activations