INDEX

Explanations

words associated with significant actions or transformations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

ght

-0.16

OGLE

-0.15

ral

-0.15

aggregate

-0.15

-0.14

 Overnight

-0.14

ty

-0.14

 Î§Î±

-0.14

bau

-0.14

POSITIVE LOGITS

Äįet

0.16

 Champion

0.15

 Diet

0.15

jian

0.14

uta

0.14

Ð²Ð°Ð¶

0.14

.BL

0.14

)((((

0.14

 Vand

0.14

heit

0.13

Activations Density 0.002%

words associated with significant actions or transformations

No Comments

No Known Activations