INDEX

Explanations

variations of the verb "to be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

fds

-0.16

Mu

-0.15

fv

-0.15

stoff

-0.14

ittest

-0.14

/oct

-0.14

pecial

-0.13

_Column

-0.13

ABCDE

-0.13

tees

-0.13

POSITIVE LOGITS

 able

0.18

avl

0.17

0.16

 willing

0.15

ulist

0.15

 literal

0.14

amat

0.14

 responsible

0.14

 explicit

0.14

jac

0.14

Activations Density 0.216%

variations of the verb "to be."

No Comments

No Known Activations