INDEX

Explanations

references to television shows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 '\''

-0.15

hawks

-0.14

elden

-0.14

ãĤıãģĽ

-0.13

å¯¸

-0.13

shaw

-0.13

yster

-0.13

Î½Î¿Ïį

-0.13

ionario

-0.12

 VÅ¡ech

-0.12

POSITIVE LOGITS

Â«

0.17

ourse

0.17

 Â«

0.16

ozor

0.14

et

0.14

oscope

0.14

Ø§Ø¨ÙĤ

0.14

åĭ¢

0.14

dech

0.13

rum

0.13

Activations Density 0.000%

references to television shows

No Comments

No Known Activations