INDEX

Explanations

phrases indicating subjects or entities that are regarded highly or favorably

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

å´İ

-0.16

stm

-0.15

-lnd

-0.15

Ð°Ð½Ð¸ÑĨ

-0.15

ergus

-0.15

æ±Ĥ

-0.15

avl

-0.15

slideDown

-0.15

iferay

-0.14

Ð°Ð»Ñĥ

-0.14

POSITIVE LOGITS

erm

0.15

 automatically

0.15

HL

0.15

 COMPLETE

0.15

 completely

0.14

 McKay

0.14

 overall

0.14

reira

0.14

birth

0.13

ãĥ³

0.13

Activations Density 0.020%

phrases indicating subjects or entities that are regarded highly or favorably

No Comments

No Known Activations