INDEX

Explanations

the letter 'a' 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 @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ager

-0.15

zia

-0.14

elle

-0.14

ling

-0.14

ced

-0.14

Ã²

-0.14

fmt

-0.13

den

-0.13

oen

-0.13

POSITIVE LOGITS

alley

0.20

à¹Ģà¸¡

0.15

lse

0.15

Ã©ru

0.14

istol

0.14

acific

0.14

ÏĦÏīÎ½

0.14

ä¸Ķ

0.13

ulty

0.13

sand

0.13

Activations Density 0.007%

the letter 'a' in various contexts

No Comments

No Known Activations