INDEX

Explanations

phrases that discuss the effectiveness or utility of actions or items

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

hurst

-0.19

anca

-0.17

rides

-0.16

sss

-0.16

achuset

-0.15

emer

-0.15

simp

-0.14

åºŃ

-0.14

ismic

-0.14

engers

-0.14

POSITIVE LOGITS

alis

0.16

lev

0.15

peg

0.14

vil

0.14

 Gregory

0.14

eti

0.14

ÄĲT

0.14

antal

0.13

eth

0.13

adian

0.13

Activations Density 0.064%

phrases that discuss the effectiveness or utility of actions or items

No Comments

No Known Activations