INDEX

Explanations

adverbs indicating gradual change or progression

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

hung

-0.16

çĶĺ

-0.15

fst

-0.14

eria

-0.14

arme

-0.13

ë¦Ħ

-0.13

HEMA

-0.13

enan

-0.13

 znaÄį

-0.13

ordion

-0.13

POSITIVE LOGITS

 slowly

0.26

 dáº§n

0.22

 gradually

0.22

 increment

0.20

 Slow

0.20

slow

0.19

 gradual

0.19

Slow

0.18

 incremental

0.17

 slow

0.16

Activations Density 0.043%

adverbs indicating gradual change or progression

No Comments

No Known Activations