INDEX

Explanations

numerical ratings and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ãĤīãģĽ

-0.16

 fours

-0.14

ä¸īä¸ī

-0.14

æı®

-0.14

utsch

-0.14

cplusplus

-0.13

argout

-0.13

 KÄ±z

-0.13

_TRUNC

-0.13

å¡ļ

-0.13

POSITIVE LOGITS

0.98

ten

0.85

Ten

0.85

TEN

0.76

Ten

0.75

åįģ

0.73

_ten

0.72

ten

0.71

 tenth

0.71

Û±Û°

0.64

Activations Density 0.140%

numerical ratings and comparisons

No Comments

No Known Activations