INDEX

Explanations

programming-related syntax and code structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

PI

-0.15

uir

-0.14

å½¦

-0.14

REC

-0.14

Pi

-0.14

ëĵł

-0.13

uc

-0.13

 Schl

-0.13

umann

-0.13

POSITIVE LOGITS

 return

0.57

return

0.50

	return

0.44

 Return

0.41

-return

0.39

 RETURN

0.37

 returning

0.37

Return

0.37

 returns

0.36

 returned

0.36

Activations Density 0.188%

programming-related syntax and code structure

No Comments

No Known Activations