INDEX

Explanations

phrases related to expectations and comparisons in quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

oksen

-0.19

ilo

-0.17

ogen

-0.15

ilon

-0.15

anz

-0.15

oust

-0.14

 shouldn

-0.14

å°¤

-0.14

ivor

-0.13

afort

-0.13

POSITIVE LOGITS

 elsewhere

0.25

0.20

 seen

0.19

 would

0.19

 fare

0.19

 similarly

0.19

 ê·¸ëłĩ

0.17

seen

0.17

 experience

0.17

 used

0.17

Activations Density 0.118%

phrases related to expectations and comparisons in quality

No Comments

No Known Activations