INDEX

Explanations

words related to lists, rankings, and recommendations

oai_token-act-pair · gpt-3.5-turbo

references to films and related categories such as sites, projects, and keywords

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pter

-0.67

roth

-0.65

atorium

-0.65

bah

-0.65

raq

-0.63

nee

-0.63

 Dhabi

-0.63

grad

-0.61

rontal

-0.60

ople

-0.60

POSITIVE LOGITS

eries

1.00

 vying

0.95

 grouped

0.84

 individually

0.70

earch

0.70

 listed

0.69

ource

0.68

 spawned

0.67

 competed

0.67

 competing

0.67

Activations Density 0.439%

words related to lists, rankings, and recommendations

references to films and related categories such as sites, projects, and keywords

No Comments

No Known Activations