INDEX

Explanations

terminology and concepts related to design

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

apon

-0.16

ystal

-0.15

ÑģÑı

-0.15

idade

-0.15

oggler

-0.15

-sized

-0.15

ht

-0.14

lick

-0.14

izona

-0.14

ordin

-0.14

POSITIVE LOGITS

able

0.18

ers

0.18

wright

0.17

/color

0.16

/colors

0.16

ees

0.15

ä¸ĢåĮº

0.15

ableObject

0.14

ABLE

0.14

ableView

0.14

Activations Density 0.036%

terminology and concepts related to design

No Comments

No Known Activations