INDEX

Explanations

phrases indicating potential utility or effectiveness of subjects and actions

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

Supported

-0.14

 Supported

-0.13

 Covered

-0.13

URRE

-0.13

Ã¸d

-0.12

iÃ³n

-0.12

ignon

-0.12

 Received

-0.12

 æ¼

-0.12

à¸´à¹ī

-0.11

POSITIVE LOGITS

 used

0.60

used

0.50

 utilized

0.48

 employed

0.48

 Used

0.45

-used

0.44

Used

0.42

employed

0.42

 USED

0.40

_used

0.40

Activations Density 0.211%

phrases indicating potential utility or effectiveness of subjects and actions

No Comments

No Known Activations