INDEX

Explanations

references to time, particularly the word "the" in relation to time periods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

urv

-0.17

STS

-0.15

struments

-0.15

 earliest

-0.15

@student

-0.15

 Sylv

-0.14

 kostenlose

-0.14

 NSLocalizedString

-0.14

 tallest

-0.14

 Zaman

-0.14

POSITIVE LOGITS

 past

0.30

 course

0.26

past

0.18

course

0.18

 proceeding

0.18

 coming

0.17

 span

0.17

è¿ĩåİ»

0.17

 Course

0.16

 years

0.16

Activations Density 0.076%

references to time, particularly the word "the" in relation to time periods

No Comments

No Known Activations