INDEX

Explanations

words and phrases that indicate relationships 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 @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ko

-0.15

ique

-0.15

istra

-0.15

dech

-0.14

croft

-0.14

undle

-0.14

onga

-0.13

 gren

-0.13

FK

-0.13

icas

-0.13

POSITIVE LOGITS

two

0.17

 sexes

0.17

 jadx

0.16

ä¸įåĲĮ

0.16

 dois

0.16

 different

0.15

ERSHEY

0.15

Ã¡r

0.15

Delimiter

0.15

ableView

0.14

Activations Density 0.182%

words and phrases that indicate relationships and comparisons

No Comments

No Known Activations