INDEX

Explanations

phrases that indicate additional information or context, particularly those beginning with "aside."

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

tÃŃ

-0.16

onica

-0.16

efeller

-0.15

UAL

-0.15

UIL

-0.14

esda

-0.14

iedy

-0.14

issy

-0.14

oad

-0.14

undi

-0.14

POSITIVE LOGITS

0.16

compose

0.15

Bes

0.14

 principals

0.14

ollow

0.14

ToObject

0.14

mad

0.14

itch

0.14

arrant

0.14

rott

0.13

Activations Density 0.007%

phrases that indicate additional information or context, particularly those beginning with "aside."

No Comments

No Known Activations