INDEX

Explanations

occurrences of the word "where" and related conditional phrases in mathematical or scientific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

vinces

-0.17

NCY

-0.15

ycastle

-0.15

çĤİ

-0.15

REATED

-0.15

Ã±ana

-0.14

naÄįenÃŃ

-0.14

addtogroup

-0.14

.charset

-0.14

 Kane

-0.14

POSITIVE LOGITS

apas

0.14

 Integr

0.14

###

0.14

uali

0.14

ì¡±

0.13

 Fully

0.13

cai

0.13

Ear

0.13

enda

0.13

Activations Density 0.024%

occurrences of the word "where" and related conditional phrases in mathematical or scientific contexts

No Comments

No Known Activations