INDEX

Explanations

references to specific lakes and associated geographical features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 brick

-0.15

jure

-0.14

 Mile

-0.14

wet

-0.14

lich

-0.14

 Millet

-0.14

endoza

-0.13

 filib

-0.13

POSITIVE LOGITS

alnÄ±z

0.16

bed

0.15

_utilities

0.15

dÄ±ÄŁÄ±nda

0.14

hammer

0.14

oh

0.14

ë²Į

0.14

hop

0.14

oleÄį

0.14

centage

0.14

Activations Density 0.046%

references to specific lakes and associated geographical features

No Comments

No Known Activations