INDEX

Explanations

references to relationships or comparisons between concepts or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ride

-0.18

uality

-0.17

ter

-0.17

elle

-0.16

rid

-0.16

th

-0.16

ess

-0.15

anova

-0.15

æĸĻ

-0.15

ra

-0.15

POSITIVE LOGITS

 humidity

0.23

 newcomer

0.20

 merits

0.19

 importance

0.19

Hum

0.18

Clause

0.17

OLUTE

0.17

 olarak

0.17

humidity

0.17

ley

0.16

Activations Density 0.022%

references to relationships or comparisons between concepts or entities

No Comments

No Known Activations