INDEX

Explanations

comparative phrases indicating preferable options or choices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

uitka

-0.15

acht

-0.15

utex

-0.15

omencl

-0.14

arness

-0.14

amage

-0.14

illac

-0.14

metro

-0.14

 primaryKey

-0.14

.xtext

-0.14

POSITIVE LOGITS

å¾

0.15

gg

0.15

zÄħ

0.14

 hands

0.14

VAS

0.14

way

0.14

nor

0.14

efs

0.14

 Pound

0.13

NG

0.13

Activations Density 0.021%

comparative phrases indicating preferable options or choices

No Comments

No Known Activations