INDEX

Explanations

references to newsletters and publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

/high

-0.16

ods

-0.14

 portion

-0.14

arhus

-0.14

ipp

-0.14

å®¹

-0.13

mas

-0.13

ãģ¼

-0.13

_dual

-0.13

Æ°a

-0.13

POSITIVE LOGITS

olare

0.16

 magazine

0.15

èªĮ

0.15

rana

0.15

ospace

0.14

 Magazine

0.14

fragistics

0.14

sheet

0.14

-ng

0.14

Äįel

0.14

Activations Density 0.088%

references to newsletters and publications

No Comments

No Known Activations