INDEX

Explanations

noun phrases indicating various levels of importance or quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

verture

-0.16

ama

-0.15

 findings

-0.15

ombs

-0.15

 prematurely

-0.14

ivery

-0.14

ieren

-0.13

 tÆ°

-0.13

 OTHERWISE

-0.13

 stance

-0.13

POSITIVE LOGITS

 reason

0.25

 chance

0.24

 possibility

0.24

 saying

0.23

 limit

0.23

 need

0.21

 difference

0.21

 temptation

0.21

lot

0.20

 danger

0.20

Activations Density 0.084%

noun phrases indicating various levels of importance or quality

No Comments

No Known Activations