INDEX

Explanations

statements indicating belief or reporting assertions

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

pee

-0.14

ucz

-0.14

aire

-0.14

ther

-0.14

Ñ

-0.14

trieve

-0.13

ieg

-0.13

porto

-0.13

.xaml

-0.13

ston

-0.13

POSITIVE LOGITS

to

0.20

-Sah

0.16

/umd

0.16

AINED

0.15

ly

0.14

'((

0.14

/arch

0.14

ivol

0.14

 Î½Î±

0.14

Activations Density 0.082%

statements indicating belief or reporting assertions

No Comments

No Known Activations