INDEX

Explanations

instances of numerical references or data values

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

zs

-0.16

orado

-0.15

RAINT

-0.15

/commons

-0.15

_TestCase

-0.15

Qed

-0.14

foy

-0.14

panies

-0.14

nodoc

-0.14

 traces

-0.14

POSITIVE LOGITS

arming

0.16

ackers

0.15

 trough

0.14

Ð°ÑĤÐµÐ»ÑĮÐ½Ð¾

0.14

 Windsor

0.13

Ð»Ð¸ÑĤ

0.13

Ø®ÙĦ

0.13

period

0.13

 misrepresented

0.13

 thru

0.13

Activations Density 0.002%

instances of numerical references or data values

No Comments

No Known Activations