INDEX

Explanations

words and phrases related to defects and imperfections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

vinc

-0.16

ury

-0.15

uron

-0.15

endir

-0.14

scriber

-0.13

 Teknik

-0.13

ghi

-0.13

erif

-0.13

ANJI

-0.13

azi

-0.13

POSITIVE LOGITS

0.18

rens

0.18

/loose

0.17

lessly

0.16

chan

0.16

ities

0.15

chers

0.15

 flawed

0.15

/conf

0.14

/problem

0.14

Activations Density 0.058%

words and phrases related to defects and imperfections

No Comments

No Known Activations