INDEX

Explanations

quoted dialogue within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

uffs

-0.20

ÑĭÑģ

-0.17

annis

-0.17

ustr

-0.15

fax

-0.14

ese

-0.14

Oro

-0.14

aty

-0.14

ury

-0.14

 reference

-0.14

POSITIVE LOGITS

istique

0.16

@js

0.16

umi

0.15

abbo

0.15

jong

0.15

importe

0.15

íĦ´

0.14

aÅŁ

0.14

ONGL

0.14

eum

0.14

Activations Density 0.022%

quoted dialogue within the text

No Comments

No Known Activations