INDEX

Explanations

instances of dates, specifically those formatted with a numerical day followed by "st" or "nd"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

Browse

-0.15

imes

-0.15

erule

-0.15

ugh

-0.14

 burden

-0.14

eres

-0.14

ething

-0.14

etus

-0.13

POSITIVE LOGITS

soever

0.17

fold

0.15

bish

0.14

ovah

0.14

éĢīæĭ

0.14

rd

0.14

ondere

0.14

å·±

0.14

unto

0.14

SGlobal

0.14

Activations Density 0.025%

instances of dates, specifically those formatted with a numerical day followed by "st" or "nd"

No Comments

No Known Activations