INDEX

Explanations

occurrences of justified legal and financial terminology

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

 CACHE

-0.15

Æ°á»¡ng

-0.15

á»Ŀi

-0.15

Ð½Ð°Ð²

-0.15

RITE

-0.15

EventManager

-0.15

unday

-0.15

 Bilim

-0.15

erte

-0.14

SizeMode

-0.14

POSITIVE LOGITS

ife

0.15

ifax

0.15

itere

0.15

 Fault

0.14

ered

0.14

 Benton

0.14

ofilm

0.14

erb

0.14

 fault

0.14

 labeling

0.14

Activations Density 0.001%

occurrences of justified legal and financial terminology

No Comments

No Known Activations