INDEX

Explanations

references to legal cases and terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ä¿Ĺ

-0.17

haar

-0.16

åĵ

-0.15

orgh

-0.15

oine

-0.15

empor

-0.14

ilyn

-0.14

ä¼į

-0.14

.pair

-0.14

zar

-0.14

POSITIVE LOGITS

ÙĦÙģ

0.15

.recv

0.14

 betray

0.14

TP

0.13

 EXEMPLARY

0.13

ako

0.13

tf

0.13

 Slice

0.13

ipay

0.13

Activations Density 0.059%

references to legal cases and terminology

No Comments

No Known Activations