INDEX

Explanations

specific numerical codes or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

Ð³Ð°Ð»Ñĸ

-0.17

Ð½Ð°Ð´Ð»ÐµÐ¶

-0.16

wo

-0.15

wi

-0.15

wa

-0.15

 span

-0.15

leck

-0.15

åĿĽ

-0.15

one

-0.14

oro

-0.14

POSITIVE LOGITS

ÐµÑħ

0.19

ÐµÐ·

0.19

Ð¸Ð³

0.19

ÑĢÐ¾Ñģ

0.18

ÑĤÑı

0.17

Ð»ÐµÑĤ

0.17

levant

0.16

ĭ

0.16

ÑĤÐ°

0.16

.scalablytyped

0.15

Activations Density 0.013%

specific numerical codes or identifiers

No Comments

No Known Activations