INDEX

Explanations

patterns of ellipses and scores of 4 and above, indicating significance or emphasis in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ereal

-0.17

ployment

-0.16

avo

-0.15

ubo

-0.14

ssue

-0.14

ayar

-0.14

ÑĢÐ°Ð²Ð¸

-0.14

allo

-0.14

aimassage

-0.14

POSITIVE LOGITS

iversit

0.15

urs

0.15

 distance

0.15

ifer

0.14

stanov

0.14

ButtonModule

0.14

unes

0.13

Å¡t

0.13

Spatial

0.13

 because

0.13

Activations Density 0.085%

patterns of ellipses and scores of 4 and above, indicating significance or emphasis in the text

No Comments

No Known Activations