INDEX

Explanations

instances of significant nouns and their attributes

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

terior

-0.15

ilk

-0.15

ãĤ¤ãĤ¹

-0.14

irt

-0.14

 Resident

-0.13

 standards

-0.13

etta

-0.13

iences

-0.13

ien

-0.13

ugar

-0.13

POSITIVE LOGITS

ayo

0.15

/moment

0.15

ismet

0.15

hora

0.15

agli

0.15

ibold

0.14

InstanceState

0.14

Î±Î½

0.14

aln

0.14

yms

0.14

Activations Density 0.041%

instances of significant nouns and their attributes

No Comments

No Known Activations