INDEX

Explanations

references to specific locations or addresses

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

Ñİ

-0.17

ric

-0.15

Ã

-0.15

zan

-0.15

-0.14

ãĥ¼ãĤ¸

-0.14

_Private

-0.14

ÑİÑĢ

-0.14

 ëĬ

-0.14

reo

-0.14

POSITIVE LOGITS

lli

0.19

uzzi

0.15

oeff

0.15

 heads

0.15

 Heads

0.14

 Beaut

0.14

eli

0.14

ERG

0.14

cab

0.14

iale

0.14

Activations Density 0.274%

references to specific locations or addresses

No Comments

No Known Activations