INDEX

Explanations

references to journal articles and their accessibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

inke

-0.18

@}

-0.15

auer

-0.15

Bro

-0.15

kont

-0.14

 Strat

-0.14

rok

-0.14

è²ł

-0.14

Å¼

-0.14

 burn

-0.14

POSITIVE LOGITS

eyim

0.17

ichick

0.16

esis

0.15

posable

0.14

ystone

0.14

OWL

0.14

oldt

0.14

 Tong

0.14

tep

0.14

ues

0.13

Activations Density 0.175%

references to journal articles and their accessibility

No Comments

No Known Activations