INDEX

Explanations

phrases indicating ease or difficulty in performing actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

inker

-0.17

oslav

-0.15

tek

-0.15

 Strap

-0.15

 ÏĦÏģÏĮ

-0.14

DEFINE

-0.14

chestra

-0.14

kova

-0.14

 å°ĳ

-0.14

ç·Ĵ

-0.14

POSITIVE LOGITS

vore

0.16

 Campos

0.16

atab

0.15

MASK

0.15

Å¡ak

0.14

ipeg

0.14

.ErrorCode

0.14

iras

0.13

Jay

0.13

Activations Density 0.032%

phrases indicating ease or difficulty in performing actions

No Comments

No Known Activations