INDEX

Explanations

references to common behavioral patterns and statistics surrounding individual experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

Ð»ÐµÐ¶

-0.16

les

-0.15

utra

-0.14

Ä°Y

-0.14

à¸Ķà¸³

-0.14

enha

-0.14

ç¤

-0.13

ordon

-0.13

uld

-0.13

slaught

-0.13

POSITIVE LOGITS

 phenomenon

0.18

 across

0.15

 æ³

0.15

/documentation

0.15

 among

0.15

lyn

0.14

 phenomena

0.14

à¥įà¤¯à¤¤

0.13

ÑĸÑĪ

0.13

Activations Density 0.167%

references to common behavioral patterns and statistics surrounding individual experiences

No Comments

No Known Activations