INDEX

Explanations

Negative statements about performance or quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

 reasonably

-0.16

utsche

-0.16

 richer

-0.15

=size

-0.15

ONGO

-0.15

Î³Î³

-0.15

 finally

-0.14

æ´¥

-0.14

ëŁŃ

-0.14

reff

-0.14

POSITIVE LOGITS

 pleasant

0.19

 impressed

0.17

 thrilled

0.17

 particularly

0.17

 ideal

0.16

 pleased

0.16

ideal

0.16

 bueno

0.15

 fond

0.15

-friendly

0.14

Activations Density 0.222%

Negative statements about performance or quality

No Comments

No Known Activations