INDEX

Explanations

phrases that indicate methods or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 store

-0.14

lico

-0.14

 ware

-0.14

 Ware

-0.14

abor

-0.14

inand

-0.14

dim

-0.13

ÏĥÏĦÎ¿

-0.13

store

-0.13

utron

-0.13

POSITIVE LOGITS

ecial

0.17

buz

0.16

ëĵ¯

0.15

Ã©ry

0.15

 Conditioning

0.14

oblin

0.14

/us

0.14

ispers

0.14

Ø§Ø¯Ø§

0.14

ajs

0.14

Activations Density 0.010%

phrases that indicate methods or processes

No Comments

No Known Activations