INDEX

Explanations

sequences of numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 IPPROTO

-0.15

ronym

-0.15

****************************************************************************

-0.15

uchen

-0.14

illaume

-0.14

ermann

-0.14

á»ĳt

-0.14

ÑħÐ¾Ð´Ð¸Ð¼

-0.14

ounty

-0.13

conde

-0.13

POSITIVE LOGITS

aneous

0.14

.gf

0.14

-translate

0.14

iese

0.14

peÄį

0.13

ONSE

0.13

 Pascal

0.13

bonus

0.13

.Flags

0.13

å½ķ

0.13

Activations Density 0.030%

sequences of numerical values

No Comments

No Known Activations