INDEX

Explanations

phrases indicating value or importance

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

otti

-0.17

zure

-0.16

istrate

-0.15

iola

-0.15

zial

-0.15

ccess

-0.15

-0.14

izada

-0.14

apolis

-0.14

/or

-0.14

POSITIVE LOGITS

iness

0.23

ful

0.21

fully

0.21

FUL

0.18

INGTON

0.17

ries

0.16

fulness

0.16

iest

0.16

ington

0.15

rops

0.15

Activations Density 0.028%

phrases indicating value or importance

No Comments

No Known Activations

phrases indicating value or importance

No Comments

No Known Activations