INDEX

Explanations

specific terms, definitions, and glossaries related to complex concepts or jargon

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

atra

-0.15

 finished

-0.15

chrift

-0.15

elman

-0.14

fax

-0.14

otto

-0.14

.ali

-0.14

ÑĢÐ¾Ð¸Ð·

-0.14

finished

-0.14

uated

-0.14

POSITIVE LOGITS

 Used

0.15

åĭ

0.15

.infinity

0.15

Used

0.15

 terms

0.14

 USED

0.14

alach

0.14

cko

0.14

oha

0.14

 defy

0.14

Activations Density 0.062%

specific terms, definitions, and glossaries related to complex concepts or jargon

No Comments

No Known Activations