INDEX

Explanations

verbs indicating actions and developments over time

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

mage

-0.19

 OMIT

-0.16

RefCount

-0.16

anki

-0.14

rzy

-0.14

iqu

-0.14

yz

-0.14

imit

-0.14

 piger

-0.14

 omas

-0.14

POSITIVE LOGITS

now

0.17

lamaz

0.16

 again

0.16

 officially

0.15

 today

0.15

 another

0.15

sf

0.14

onga

0.14

lif

0.14

us

0.13

Activations Density 0.313%

verbs indicating actions and developments over time

No Comments

No Known Activations