INDEX

Explanations

the word "respectively."

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

-0.17

ught

-0.16

erin

-0.15

rolled

-0.15

opp

-0.15

/download

-0.14

voie

-0.14

uler

-0.14

_LEG

-0.13

POSITIVE LOGITS

aleb

0.16

.Criteria

0.15

ascending

0.14

ovol

0.14

ImageContext

0.14

Ã¤nd

0.14

ìĿ´ìĬ¤

0.14

andre

0.14

igli

0.13

andler

0.13

Activations Density 0.009%

the word "respectively."

No Comments

No Known Activations