INDEX

Explanations

questions starting with "How" and their variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

whether

-0.14

lier

-0.14

WA

-0.14

ovable

-0.14

 Whether

-0.14

ÏĢÎ¹

-0.13

aneously

-0.13

elier

-0.13

 proh

-0.13

POSITIVE LOGITS

did

0.35

 does

0.29

do

0.29

did

0.27

Did

0.22

Did

0.22

 long

0.21

.did

0.21

)did

0.21

old

0.20

Activations Density 0.036%

questions starting with "How" and their variants

No Comments

No Known Activations