INDEX

Explanations

proper nouns, particularly names and titles associated with mythology or historical figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

fts

-0.17

iner

-0.16

ÅĻe

-0.15

igit

-0.15

igo

-0.14

è¼Ŀ

-0.14

ÑĮÑİÑĤ

-0.14

èĻ

-0.14

è®¸

-0.14

POSITIVE LOGITS

ÙĬØ«

0.17

à¸Ńà¸ļ

0.15

-sidebar

0.14

ernes

0.14

unde

0.14

udic

0.14

åĨł

0.13

ogenerated

0.13

ud

0.13

car

0.13

Activations Density 0.000%

proper nouns, particularly names and titles associated with mythology or historical figures

No Comments

No Known Activations