INDEX

Explanations

references to menus and navigation options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ly

-0.17

zw

-0.16

ensen

-0.16

rias

-0.16

uments

-0.15

ument

-0.15

csi

-0.15

pu

-0.14

anium

-0.14

uzzi

-0.14

POSITIVE LOGITS

ìĸ¼

0.20

onder

0.19

item

0.19

antry

0.17

ittest

0.17

stration

0.16

yat

0.16

ifer

0.16

items

0.16

Items

0.15

Activations Density 0.014%

references to menus and navigation options

No Comments

No Known Activations