INDEX

Explanations

references to applications and tools, particularly in educational and creative contexts

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

ptype

-0.15

allah

-0.15

éļĨ

-0.15

yor

-0.14

Ð½Ð°ÑĢ

-0.14

 danmark

-0.14

 Kund

-0.13

å¤§äºº

-0.13

 Glad

-0.13

urga

-0.13

POSITIVE LOGITS

.scalablytyped

0.15

ipur

0.14

tion

0.14

ees

0.14

es

0.13

.Dial

0.13

Ã¢ce

0.13

smouth

0.12

Activations Density 0.031%

references to applications and tools, particularly in educational and creative contexts

No Comments

No Known Activations