INDEX

Explanations

phrases that indicate partnerships or collaborations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

rac

-0.16

raud

-0.16

ocator

-0.15

utto

-0.15

alles

-0.14

¥Ĳ

-0.14

ouz

-0.14

olet

-0.14

à¤¾à¤¹à¤ķ

-0.14

ocale

-0.14

POSITIVE LOGITS

 Bott

0.15

nings

0.15

cil

0.15

uzzi

0.15

ifiable

0.14

 Conrad

0.14

ê»ĺ

0.13

è«

0.13

osal

0.13

Activations Density 0.021%

phrases that indicate partnerships or collaborations

No Comments

No Known Activations