INDEX

Explanations

phrases that indicate examination or observation of subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

isz

-0.17

west

-0.15

 kháº¯c

-0.15

ÙĪÚ©

-0.15

èĿ

-0.14

orts

-0.14

onnement

-0.14

 Downs

-0.13

_hint

-0.13

plen

-0.13

POSITIVE LOGITS

asso

0.17

 Pron

0.16

amb

0.15

 pronunciation

0.14

 pron

0.14

 ëĪ

0.14

 mechan

0.13

essian

0.13

 refr

0.13

afi

0.13

Activations Density 0.016%

phrases that indicate examination or observation of subjects

No Comments

No Known Activations