INDEX

Explanations

phrases indicating uncertainty or conditions regarding influence and effectiveness

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

-0.16

cho

-0.15

Vic

-0.15

 nedir

-0.14

aster

-0.14

eter

-0.14

ãģıãģł

-0.14

ser

-0.14

ãĤ¦ãĥ³

-0.14

bro

-0.13

POSITIVE LOGITS

unsch

0.17

Å¡ak

0.16

ewire

0.15

ÐĲÑĢÑħÑĸÐ²

0.15

å®¾

0.15

istrovstvÃŃ

0.15

zee

0.14

 pill

0.14

eyse

0.14

anki

0.14

Activations Density 0.347%

phrases indicating uncertainty or conditions regarding influence and effectiveness

No Comments

No Known Activations