INDEX

Explanations

superlative adjectives indicating the highest degree of a quality or characteristic

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

 enough

-0.23

ish

-0.19

 Enough

-0.19

isher

-0.17

er

-0.16

umi

-0.15

nings

-0.15

umat

-0.15

ings

-0.14

 beyond

-0.14

POSITIVE LOGITS

-ever

0.36

ablish

0.32

-selling

0.21

possible

0.21

 possible

0.21

/sm

0.20

/fast

0.19

 ever

0.18

oft

0.18

 Possible

0.18

Activations Density 0.063%

superlative adjectives indicating the highest degree of a quality or characteristic

No Comments

No Known Activations