INDEX

Explanations

phrases that reference rewards and incentives for participation or achievement

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

igli

-0.15

vak

-0.15

ixel

-0.15

peg

-0.15

itchens

-0.15

ampo

-0.14

Fol

-0.14

ija

-0.14

uge

-0.14

Ã¦k

-0.14

POSITIVE LOGITS

 rewards

0.17

 reward

0.17

ToAdd

0.16

antino

0.15

ham

0.15

OLER

0.14

 Kris

0.14

 Rewards

0.14

ront

0.14

ivet

0.14

Activations Density 0.136%

phrases that reference rewards and incentives for participation or achievement

No Comments

No Known Activations

phrases that reference rewards and incentives for participation or achievement

No Comments

No Known Activations