INDEX

Explanations

references to medical and scientific research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

udos

-0.15

UDO

-0.15

robat

-0.15

zeich

-0.14

Ð»ÑıÑħ

-0.14

jee

-0.14

TION

-0.14

ebek

-0.14

izmet

-0.14

erguson

-0.14

POSITIVE LOGITS

Lip

0.15

 Replay

0.14

shire

0.14

ifiable

0.14

ks

0.13

ch

0.13

eee

0.13

 Prompt

0.13

anghai

0.13

 Lover

0.13

Activations Density 0.048%

references to medical and scientific research

No Comments

No Known Activations