INDEX

Explanations

proper nouns, particularly names and locations

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

oj

-0.14

 Ð¿ÐµÑĢÐµÐº

-0.14

ollision

-0.14

aben

-0.14

Rig

-0.14

irement

-0.14

-0.13

 rece

-0.13

okud

-0.13

rig

-0.13

POSITIVE LOGITS

Ìģ

0.17

hoff

0.16

ancode

0.15

¤íĶĦ

0.15

/OR

0.15

 microtime

0.14

ÄĽlÃŃ

0.14

.uml

0.13

bell

0.13

rippling

0.13

Activations Density 0.231%

proper nouns, particularly names and locations

No Comments

No Known Activations