INDEX

Explanations

terms related to improvement and evaluation in a variety of contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

paque

-0.17

 ConnectionState

-0.16

onde

-0.16

Ú¯ÛĮØ±

-0.15

FetchRequest

-0.15

ahu

-0.15

/umd

-0.14

NCY

-0.14

lexport

-0.14

nodoc

-0.14

POSITIVE LOGITS

azers

0.14

orest

0.14

 early

0.14

 Prof

0.13

 Rescue

0.13

ÏģÎ¿Ïį

0.13

fir

0.13

ÃĹ

0.13

’Ñı

0.13

onis

0.13

Activations Density 0.069%

terms related to improvement and evaluation in a variety of contexts

No Comments

No Known Activations