INDEX

Explanations

phrases emphasizing detail and precision

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

etty

-0.17

cono

-0.16

sink

-0.15

 Witness

-0.15

lasses

-0.14

Witness

-0.14

Uvs

-0.14

.paging

-0.14

hardt

-0.14

à¤¬à¤²

-0.14

POSITIVE LOGITS

.detail

0.16

ç»Ĩ

0.16

illi

0.16

alia

0.15

 detail

0.15

ç´°

0.15

.Detail

0.15

 details

0.14

 Resolution

0.14

Resolution

0.14

Activations Density 0.259%

phrases emphasizing detail and precision

No Comments

No Known Activations