INDEX

Explanations

words and phrases indicating queries or inquiries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ring

-0.15

Ã¼zel

-0.15

mps

-0.15

-ring

-0.15

 ring

-0.14

.POS

-0.14

claimer

-0.14

zad

-0.14

 subrange

-0.13

_times

-0.13

POSITIVE LOGITS

çĽ®çļĦ

0.16

 purpose

0.16

 Goal

0.15

 ÙĩØ¯Ùģ

0.15

 goal

0.15

 ÑĨÐµÐ»ÑĮ

0.15

 intention

0.15

RIORITY

0.15

 targeted

0.15

Goal

0.15

Activations Density 0.015%

words and phrases indicating queries or inquiries

No Comments

No Known Activations

words and phrases indicating queries or inquiries

No Comments

No Known Activations