INDEX

Explanations

quantifiable data and statistics

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

 Rare

-0.16

ÃŁen

-0.15

rar

-0.14

_PROXY

-0.14

omid

-0.14

ieber

-0.14

Rare

-0.14

ork

-0.14

aken

-0.13

omez

-0.13

POSITIVE LOGITS

 respectively

1.05

respect

0.87

 respective

0.73

 respect

0.70

 Respect

0.68

 ÑģÐ¾Ð¾ÑĤÐ²ÐµÑĤ

0.65

 resp

0.63

resp

0.56

,res

0.55

åĪĨåĪ«

0.54

Activations Density 0.115%

quantifiable data and statistics

No Comments

No Known Activations