INDEX

Explanations

terms related to outcomes and results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

thing

-0.18

oria

-0.18

igel

-0.17

Ø®Ø§ÙĨÙĩ

-0.17

als

-0.15

itter

-0.15

ALLOC

-0.15

../../../

-0.15

lie

-0.14

bles

-0.14

POSITIVE LOGITS

antly

0.29

ados

0.27

ë¬¼ìĿĦ

0.25

ë¬¼

0.25

ants

0.22

-oriented

0.21

ative

0.20

/result

0.20

ADOS

0.20

swith

0.19

Activations Density 0.063%

terms related to outcomes and results

No Comments

No Known Activations