INDEX

Explanations

adverbs that modify actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ÙıÙĨ

-0.15

Î¸Î¼

-0.14

 ÑĢÐ°Ð±

-0.14

akes

-0.14

kers

-0.14

-0.13

åı·

-0.13

Ð¾Ð½Ñĭ

-0.13

cate

-0.13

POSITIVE LOGITS

Ã¼rk

0.17

igh

0.16

à¹Ĩ

0.16

ahoma

0.15

_Style

0.15

 manner

0.15

 Booth

0.15

uluk

0.15

issa

0.15

/at

0.15

Activations Density 0.322%

adverbs that modify actions

No Comments

No Known Activations