INDEX

Explanations

expressions of personal opinions and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ildo

-0.18

.opendaylight

-0.17

 fitte

-0.16

Sure

-0.16

owitz

-0.16

 organizer

-0.15

sure

-0.15

ADED

-0.15

aded

-0.14

indr

-0.14

POSITIVE LOGITS

 shall

0.17

Vid

0.16

 proper

0.16

znik

0.16

ps

0.16

 broadly

0.15

proper

0.15

 Shall

0.15

reck

0.15

 spoken

0.15

Activations Density 0.140%

expressions of personal opinions and experiences

No Comments

No Known Activations

expressions of personal opinions and experiences

No Comments

No Known Activations