INDEX

Explanations

phrases indicating difficulty or challenges

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 ZÃ¡p

-0.15

 Deprecated

-0.14

 Barg

-0.14

impan

-0.14

 seriousness

-0.13

DataAdapter

-0.13

eker

-0.13

 Deng

-0.12

_warnings

-0.12

ollen

-0.12

POSITIVE LOGITS

 difficult

0.82

 hard

0.73

 harder

0.71

 hardest

0.65

 diffic

0.62

 difficulty

0.60

hard

0.59

 difÃŃcil

0.59

-hard

0.58

diff

0.58

Activations Density 0.341%

phrases indicating difficulty or challenges

No Comments

No Known Activations