INDEX

Explanations

references to websites, blogs, and internet-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 Snape

-0.15

¥Ĳ

-0.15

andel

-0.14

Å©i

-0.14

[section

-0.14

 Sovere

-0.14

SKU

-0.14

 spherical

-0.13

lename

-0.13

sph

-0.13

POSITIVE LOGITS

 site

1.04

-site

0.87

 Site

0.85

 sites

0.84

site

0.83

Site

0.78

_site

0.76

 SITE

0.71

 Sites

0.69

ãĤµãĤ¤ãĥĪ

0.66

Activations Density 0.192%

references to websites, blogs, and internet-related activities

No Comments

No Known Activations