INDEX

Explanations

phrases indicating actions, instructions, or recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 Leban

-0.17

æ¡ĥ

-0.16

PackageName

-0.15

/GL

-0.14

REFER

-0.14

TouchUpInside

-0.14

uye

-0.14

unya

-0.14

RAP

-0.14

OperationException

-0.13

POSITIVE LOGITS

0.17

obus

0.16

which

0.15

 which

0.15

ä¾Ľ

0.15

aji

0.15

BI

0.14

 accompany

0.14

ilian

0.14

MESS

0.13

Activations Density 0.183%

phrases indicating actions, instructions, or recommendations

No Comments

No Known Activations