INDEX

Explanations

terms related to structure and organization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

aira

-0.16

Ã¶m

-0.15

 Cooke

-0.14

rese

-0.14

iences

-0.13

Ãºc

-0.13

isser

-0.13

 Airways

-0.13

house

-0.13

 Bull

-0.13

POSITIVE LOGITS

elon

0.18

uali

0.18

gorm

0.17

0.16

AccessException

0.15

 Ð³Ð°Ð»

0.15

Ngh

0.15

-prepend

0.15

alli

0.15

 Synthetic

0.14

Activations Density 0.023%

terms related to structure and organization

No Comments

No Known Activations