INDEX

Explanations

references to notable individuals and technological terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

leÅŁik

-0.18

lish

-0.16

ÑĪÐ¸Ð±

-0.16

-valu

-0.16

ìķ¼

-0.16

åĭ¤

-0.15

typeorm

-0.15

ligt

-0.15

-boy

-0.15

================================================================

-0.15

POSITIVE LOGITS

ingly

0.21

nÃŃ

0.19

åĻ¨

0.18

/fire

0.18

men

0.17

ians

0.17

proof

0.17

Activations Density 0.332%

references to notable individuals and technological terms

No Comments

No Known Activations