INDEX

Explanations

phrases indicating offers of help or support

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

owy

-0.16

acro

-0.15

 Personnel

-0.15

uro

-0.15

swire

-0.15

uell

-0.14

ieri

-0.14

owe

-0.14

cs

-0.14

ast

-0.14

POSITIVE LOGITS

orthand

0.16

 Ø®Ø§Ø±Ø¬ÙĬØ©

0.14

PackageManager

0.14

essian

0.14

Ð°ÑĢÐ°Ð¼

0.14

/epl

0.14

ogens

0.13

oved

0.13

itness

0.13

Ø°Ø§Ø±

0.13

Activations Density 0.012%

phrases indicating offers of help or support

No Comments

No Known Activations