INDEX

Explanations

past tense verbs and phrases indicating completed actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

zew

-0.17

ptom

-0.15

vin

-0.15

 Neutral

-0.15

vi

-0.15

 Civil

-0.15

as

-0.15

 soon

-0.15

Vin

-0.14

Ð°ÑĤÐ¾Ð²

-0.14

POSITIVE LOGITS

 ÙħØ¨Ø§Ø´

0.17

 xong

0.16

peare

0.16

adoo

0.16

haled

0.15

onio

0.15

ayet

0.15

 adulti

0.15

ichni

0.15

Activations Density 0.210%

past tense verbs and phrases indicating completed actions

No Comments

No Known Activations