INDEX

Explanations

terms associated with measurement or evaluation

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

ocos

-0.17

ches

-0.17

udeau

-0.16

-0.14

sao

-0.14

acades

-0.14

terdam

-0.14

 digest

-0.14

uai

-0.14

 Ð¶Ðµ

-0.14

POSITIVE LOGITS

orne

0.16

ORB

0.16

ccb

0.15

weg

0.15

atem

0.14

istrovstvÃŃ

0.14

anim

0.14

 Press

0.14

sheet

0.14

ither

0.14

Activations Density 0.008%

terms associated with measurement or evaluation

No Comments

No Known Activations