INDEX

Explanations

interrogative and conditional phrases implying uncertainty or doubt

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 though

-0.15

 while

-0.15

uche

-0.15

.TODO

-0.14

æīį

-0.14

 since

-0.14

è³¢

-0.13

stup

-0.13

 after

-0.13

ÂŃt

-0.13

POSITIVE LOGITS

And

0.34

And

0.30

AndGet

0.21

 Cause

0.20

Cause

0.19

_and

0.19

So

0.18

 Number

0.18

So

0.18

-and

0.18

Activations Density 0.126%

interrogative and conditional phrases implying uncertainty or doubt

No Comments

No Known Activations