INDEX

Explanations

discussions focused on equality and non-discrimination among individuals based on various identity factors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

isque

-0.15

arges

-0.15

 actionTypes

-0.15

#ab

-0.14

Ð»Ð°Ð²

-0.14

 Ø¯Ø§Ùħ

-0.14

Ð³Ð»

-0.14

ohen

-0.14

pawn

-0.14

#aa

-0.14

POSITIVE LOGITS

 race

0.58

race

0.47

 Race

0.44

Race

0.42

 races

0.41

 religion

0.40

 ethnicity

0.39

_race

0.36

 nationality

0.34

 Races

0.34

Activations Density 0.250%

discussions focused on equality and non-discrimination among individuals based on various identity factors

No Comments

No Known Activations