INDEX

Explanations

references to historical artifacts and archaeological findings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

antan

-0.15

 Cabinets

-0.14

.ogg

-0.14

 titan

-0.13

.md

-0.13

 Kemal

-0.13

arel

-0.13

 Scientists

-0.13

 literal

-0.13

 dirig

-0.13

POSITIVE LOGITS

 stylist

0.17

 Finds

0.16

 finds

0.15

ipes

0.15

 uten

0.14

 moda

0.14

çĮª

0.14

 Weinstein

0.14

 Horizon

0.14

Ã³j

0.14

Activations Density 0.052%

references to historical artifacts and archaeological findings

No Comments

No Known Activations