INDEX

Explanations

the word "just" in various contexts

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

 Blitz

-0.17

SER

-0.15

-0.14

fal

-0.14

enz

-0.14

asse

-0.14

ï¿

-0.13

tright

-0.13

ieu

-0.13

ldb

-0.13

POSITIVE LOGITS

 Ø¨ÙĦÚ©Ùĩ

0.17

mere

0.16

 LIMITED

0.16

cth

0.15

 Ekon

0.15

ccione

0.15

ermal

0.14

aty

0.14

dee

0.14

imizer

0.14

Activations Density 0.019%

the word "just" in various contexts

No Comments

No Known Activations