INDEX

Explanations

terms related to printers and printing technology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 poly

-0.16

 Cant

-0.15

dro

-0.15

 Poly

-0.15

ansk

-0.14

olare

-0.14

dro

-0.14

 Addiction

-0.13

Routing

-0.13

oling

-0.13

POSITIVE LOGITS

su

0.16

IFO

0.15

Ð¾ÑĢÐ¾Ð½

0.15

adier

0.15

kop

0.14

ezi

0.14

papers

0.14

-svg

0.14

Ø§Ø°

0.14

èħ¿

0.14

Activations Density 0.010%

terms related to printers and printing technology

No Comments

No Known Activations