INDEX

Explanations

phrases expressing effort and hard work

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ãĤ¹ãĤ¿ãĥ¼

-0.15

unken

-0.15

ritten

-0.15

 Holt

-0.15

esser

-0.15

gency

-0.14

stuff

-0.14

ucas

-0.14

rage

-0.14

HELP

-0.13

POSITIVE LOGITS

 hard

0.58

hard

0.47

 harder

0.45

 HARD

0.42

-hard

0.42

 Hard

0.39

Hard

0.38

_hard

0.37

 hardest

0.37

ç¡¬

0.36

Activations Density 0.094%

phrases expressing effort and hard work

No Comments

No Known Activations