INDEX

Explanations

technical terms and parameters related to programming or data processing

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

iros

-0.16

ogra

-0.16

bon

-0.16

 Gone

-0.15

 Noon

-0.15

imoto

-0.15

ban

-0.14

 Capac

-0.14

esel

-0.14

quir

-0.14

POSITIVE LOGITS

åĲ¦

0.17

vi

0.16

UED

0.15

ãĤ¶ãĥ¼

0.15

-Cs

0.15

&action

0.15

ì³

0.14

ASA

0.14

Activations Density 0.002%

technical terms and parameters related to programming or data processing

No Comments

No Known Activations