INDEX

Explanations

keywords and actions related to program features and functionality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ãĥķãĥĪ

-0.17

greSQL

-0.16

 ÙħÛĮÙĦØ§Ø¯ÛĮ

-0.16

ab

-0.15

led

-0.15

hamster

-0.15

Dot

-0.15

ft

-0.15

mar

-0.14

âĢª

-0.14

POSITIVE LOGITS

0.25

ING

0.21

tember

0.21

 acronym

0.18

LY

0.18

tv

0.18

ngine

0.18

VILLE

0.18

/OR

0.18

scri

0.17

Activations Density 0.265%

keywords and actions related to program features and functionality

No Comments

No Known Activations