INDEX

Explanations

phrases indicating actions or relationships involving movement or transition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÐºÐ¾ÑĤ

-0.15

 annonces

-0.14

_EOL

-0.14

 selber

-0.14

RuleContext

-0.13

OffsetTable

-0.13

ĶåĽŀ

-0.13

.PerformLayout

-0.13

eless

-0.13

edback

-0.13

POSITIVE LOGITS

the

0.28

 those

0.26

 some

0.26

 what

0.25

how

0.22

 both

0.21

 certain

0.21

 various

0.21

0.20

the

0.20

Activations Density 1.644%

phrases indicating actions or relationships involving movement or transition

No Comments

No Known Activations