INDEX

Explanations

references to strategy games, particularly the Total War series and related elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ÎºÎŃ

-0.14

 Clarence

-0.14

odo

-0.13

ÄŁa

-0.13

imit

-0.13

Ø¨Ø±

-0.13

borough

-0.13

åį

-0.13

elan

-0.13

Gor

-0.13

POSITIVE LOGITS

ald

0.15

ults

0.15

ouz

0.15

maz

0.14

antz

0.14

 ÙģØ§Ø±Ø³

0.14

arsity

0.13

trys

0.13

owers

0.13

Maz

0.13

Activations Density 0.019%

references to strategy games, particularly the Total War series and related elements

No Comments

No Known Activations