INDEX

Explanations

components and interactions related to chat systems and information retrieval

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

vetica

-0.15

Ð¾Ð²Ð¾ÑĢ

-0.15

 LETTER

-0.14

 Morrow

-0.14

 caut

-0.13

avir

-0.13

/vnd

-0.13

mÃ¼r

-0.13

 League

-0.13

 trait

-0.13

POSITIVE LOGITS

 entity

0.24

 span

0.23

 Span

0.23

 entities

0.23

 Entities

0.23

 passage

0.22

 mention

0.22

 spans

0.22

 answer

0.21

 Passage

0.21

Activations Density 0.010%

components and interactions related to chat systems and information retrieval

No Comments

No Known Activations