INDEX

Explanations

numerical values, particularly a significant one

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 bubb

-0.17

PRINTF

-0.16

urga

-0.15

Ð¾

-0.15

urg

-0.15

iri

-0.14

roy

-0.14

es

-0.14

ystone

-0.14

 fork

-0.14

POSITIVE LOGITS

 Trident

0.15

tains

0.14

 preferredStyle

0.14

ãĥ³ãĥĶ

0.14

uali

0.14

 ÑģÐ¼Ð¾ÑĤ

0.14

ãĤ¦ãĤ¹

0.14

 Tamb

0.14

Ð¾Ð²Ð¸Ñĩ

0.13

 FileMode

0.13

Activations Density 0.000%

numerical values, particularly a significant one

No Comments

No Known Activations