INDEX

Explanations

numerical and structural details associated with physical measurements

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

IMIT

-0.17

ekler

-0.16

ëŀ

-0.16

anford

-0.16

imus

-0.15

ekl

-0.15

andex

-0.14

Ø±ÙĪØª

-0.14

åŁ¹

-0.14

POSITIVE LOGITS

MB

0.18

Bol

0.17

MB

0.17

mb

0.17

Bro

0.17

Bro

0.17

Ã³ng

0.16

 Brooks

0.16

bro

0.16

_lineno

0.16

Activations Density 0.035%

numerical and structural details associated with physical measurements

No Comments

No Known Activations