INDEX

Explanations

phrases related to understanding and evaluating processes or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

yen

-0.13

Ã¤ge

-0.12

 eiusmod

-0.12

soever

-0.12

à¸¢

-0.12

opy

-0.12

Ã¡m

-0.12

ytt

-0.12

egrity

-0.12

.fi

-0.11

POSITIVE LOGITS

how

0.79

why

0.61

 what

0.57

how

0.56

 where

0.55

 whether

0.51

 cÃ³mo

0.45

å¦Ĥä½ķ

0.44

 exactly

0.43

what

0.42

Activations Density 0.663%

phrases related to understanding and evaluating processes or concepts

No Comments

No Known Activations