INDEX

Explanations

terms related to recognition and appreciation

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

elman

-0.17

forme

-0.16

>Main

-0.16

iggs

-0.16

aight

-0.15

ermo

-0.15

arch

-0.15

ARCH

-0.14

kul

-0.14

lak

-0.14

POSITIVE LOGITS

borg

0.16

VOKE

0.15

'gc

0.14

ovo

0.14

 createSelector

0.14

ac

0.14

/assert

0.14

opies

0.13

zon

0.13

orc

0.13

Activations Density 0.050%

terms related to recognition and appreciation

No Comments

No Known Activations