INDEX

Explanations

HTML or code formatting elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

ear

-0.15

.trade

-0.14

oky

-0.13

ivor

-0.13

DJ

-0.13

 Mother

-0.13

 Convention

-0.13

Wid

-0.13

 Pearl

-0.13

erno

-0.13

POSITIVE LOGITS

ampa

0.17

pard

0.16

mee

0.15

brit

0.15

avra

0.14

 sense

0.14

 isize

0.13

 meas

0.13

ilon

0.13

afia

0.13

Activations Density 0.032%

HTML or code formatting elements

No Comments

No Known Activations

HTML or code formatting elements

No Comments

No Known Activations