INDEX

Explanations

conjunctions and phrases indicating contrast or addition in arguments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

-0.15

amo

-0.15

 whom

-0.15

 either

-0.15

 somewhere

-0.14

KI

-0.14

-0.13

ni

-0.13

oh

-0.13

inch

-0.13

POSITIVE LOGITS

of

0.25

á»§a

0.23

 cá»§a

0.22

/of

0.16

 especialmente

0.15

à¸Ĥà¸Ńà¸ĩ

0.15

especially

0.15

.generated

0.15

 especially

0.14

 à¸Ĥà¸Ńà¸ĩ

0.14

Activations Density 0.334%

conjunctions and phrases indicating contrast or addition in arguments

No Comments

No Known Activations

conjunctions and phrases indicating contrast or addition in arguments

No Comments

No Known Activations