INDEX

Explanations

references and citations in academic writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 scratch

-0.16

Ð²Ð¸ÑĤ

-0.16

Ð¾Ð¶Ðµ

-0.15

 scratches

-0.15

agne

-0.15

 rounds

-0.15

Gle

-0.14

 Scratch

-0.14

 Payne

-0.14

ena

-0.14

POSITIVE LOGITS

/sources

0.17

_tF

0.15

.Native

0.15

 ÑĦÑĥÐ½Ð´Ð°

0.15

oref

0.15

abbo

0.14

encil

0.14

heels

0.14

 ÑĤÐ²Ð°

0.14

unate

0.14

Activations Density 0.101%

references and citations in academic writing

No Comments

No Known Activations