INDEX

Explanations

words and phrases related to medical knowledge and research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ø§ØŃÛĮ

-0.17

æĮ¯ãĤĬ

-0.17

ÑĥÐ¶Ðµ

-0.16

edException

-0.15

_membership

-0.15

erce

-0.15

ILA

-0.15

itis

-0.14

incare

-0.14

ÑģÐ¼

-0.13

POSITIVE LOGITS

zos

0.16

inker

0.15

insk

0.14

StackSize

0.14

 ë°±

0.14

obus

0.14

 Hilton

0.14

iever

0.13

âĢ«

0.13

ortal

0.13

Activations Density 0.039%

words and phrases related to medical knowledge and research

No Comments

No Known Activations