INDEX

Explanations

phrases indicating companionship or joint experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

oves

-0.17

 Parallel

-0.15

Parallel

-0.15

tero

-0.15

ouz

-0.15

oucher

-0.15

odel

-0.14

usic

-0.14

uz

-0.14

rouw

-0.14

POSITIVE LOGITS

0.15

zte

0.15

nob

0.15

 Futures

0.14

 Bite

0.14

bu

0.13

.handleClick

0.13

getMock

0.13

DeviceInfo

0.13

Activations Density 0.146%

phrases indicating companionship or joint experiences

No Comments

No Known Activations