INDEX

Explanations

quantifiers that express magnitude or intensity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

mani

-0.16

 quite

-0.16

erten

-0.16

animate

-0.15

 Quite

-0.15

iginal

-0.15

ÑĨÐ¸Ð½

-0.15

rou

-0.15

idious

-0.14

xic

-0.14

POSITIVE LOGITS

 nearly

0.23

 easily

0.20

 completely

0.19

 completamente

0.19

 fully

0.18

asily

0.18

 Easily

0.18

ibly

0.18

Nearly

0.18

fully

0.17

Activations Density 0.121%

quantifiers that express magnitude or intensity

No Comments

No Known Activations