INDEX

Explanations

references to television shows or programs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

adier

-0.16

awi

-0.15

 gener

-0.15

Ð½ÑıÐ²

-0.14

ÐºÐ¸Ð´

-0.14

ipsis

-0.14

bs

-0.14

 multiplic

-0.14

ìĨ

-0.13

resher

-0.13

POSITIVE LOGITS

uire

0.15

ory

0.14

 Extreme

0.14

ami

0.14

pak

0.14

Ã¶y

0.13

ÙĪÙĨÛĮ

0.13

/svg

0.13

 series

0.13

 Guest

0.13

Activations Density 0.104%

references to television shows or programs

No Comments

No Known Activations