INDEX

Explanations

references to organizational divisions or categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

-0.15

ienne

-0.14

 Vaults

-0.14

Ð½Ð°ÑĤ

-0.14

ints

-0.14

ocha

-0.13

Ð»ÑĮ

-0.13

 ########.

-0.13

icie

-0.13

ÑīÐ¸Ð½

-0.13

POSITIVE LOGITS

alone

0.16

branch

0.15

illis

0.15

Ã¹

0.15

/sub

0.15

 branches

0.14

(branch

0.14

wizard

0.14

iles

0.14

bos

0.14

Activations Density 0.042%

references to organizational divisions or categories

No Comments

No Known Activations