INDEX

Explanations

proper nouns related to organizations, titles, and individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to individuals' names and titles

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tantal

-0.66

 hurdle

-0.65

isSpecialOrderable

-0.62

Advertisements

-0.61

 chilling

-0.61

.'"

-0.61

 gaps

-0.60

 Hawking

-0.60

 obstacles

-0.59

 Ukrainians

-0.59

POSITIVE LOGITS

 Village

0.67

 Industries

0.65

 Projects

0.60

 Merchants

0.60

Portland

0.59

Dialogue

0.59

chan

0.59

omics

0.59

 Corona

0.58

 Cong

0.58

Activations Density 0.530%

proper nouns related to organizations, titles, and individuals

proper nouns related to individuals' names and titles

No Comments

No Known Activations