INDEX

Explanations

phrases that indicate the credibility and seriousness of a show or investigation

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

atica

-0.15

pyx

-0.14

erialize

-0.14

achuset

-0.14

ÑĪÑĤ

-0.14

cia

-0.14

ÎŃÎ½

-0.13

pler

-0.13

éĽĨä¸Ń

-0.13

anim

-0.13

POSITIVE LOGITS

 realism

0.38

 realistic

0.36

 reality

0.28

 literal

0.25

 realistically

0.25

 Real

0.24

 Reality

0.24

 real

0.24

-real

0.24

 rÃ©al

0.23

Activations Density 0.031%

phrases that indicate the credibility and seriousness of a show or investigation

No Comments

No Known Activations