INDEX

Explanations

product model numbers and specific item identifiers

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

anja

-0.15

acey

-0.14

 unlikely

-0.14

Î¹Î±Îº

-0.13

 coal

-0.13

uckle

-0.13

eya

-0.13

egg

-0.13

ello

-0.13

POSITIVE LOGITS

Ã³g

0.17

ennent

0.15

 Briggs

0.14

 ëĭ¤ìļ´ë°Ľê¸°

0.14

shore

0.14

 ëĦ¤ìĿ´íĬ¸

0.14

istrov

0.13

.ga

0.13

.readValue

0.13

 toasted

0.13

Activations Density 0.076%

product model numbers and specific item identifiers

No Comments

No Known Activations