INDEX

Explanations

forms of the verb "to be" and their negations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

wner

-0.19

iyan

-0.19

isoft

-0.17

Ã±ana

-0.15

æĽ¼

-0.15

bject

-0.14

forman

-0.14

rahim

-0.14

ayar

-0.14

/setup

-0.14

POSITIVE LOGITS

 doing

0.20

 trying

0.17

 Doing

0.15

 saying

0.15

 bÃ½t

0.15

 making

0.14

ood

0.14

 down

0.14

angu

0.14

Activations Density 0.199%

forms of the verb "to be" and their negations in various contexts

No Comments

No Known Activations