INDEX

Explanations

references to reality TV shows and competitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

isphere

-0.19

èįī

-0.18

utto

-0.15

andex

-0.14

auga

-0.14

 Vict

-0.14

 Wins

-0.14

 counter

-0.14

èªł

-0.14

EIF

-0.13

POSITIVE LOGITS

 Season

0.17

 PageSize

0.16

 ÙħØ´Ø§Ø±

0.16

 format

0.15

rek

0.14

UK

0.14

flen

0.14

Å¡il

0.14

enci

0.14

Activations Density 0.032%

references to reality TV shows and competitions

No Comments

No Known Activations