INDEX

Explanations

phrases indicating purpose or intention

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

igua

-0.15

redient

-0.15

udos

-0.14

 Cinder

-0.14

Nun

-0.14

uego

-0.14

resse

-0.13

acje

-0.13

redients

-0.13

stial

-0.13

POSITIVE LOGITS

essen

0.17

pring

0.15

ÅĤÄħ

0.14

 subject

0.14

Ð¼ÑĸÐ½

0.14

andering

0.14

.ali

0.14

agen

0.14

¨¡

0.14

adu

0.14

Activations Density 0.012%

phrases indicating purpose or intention

No Known Activations