INDEX

Explanations

vocabularies associated with expenses and monetary value

oai_token-act-pair · gpt-3.5-turbo

references to financial expenses or costs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ZE

-0.71

watching

-0.67

here

-0.66

 =================================

-0.66

Favorite

-0.62

igel

-0.62

ocent

-0.62

âĦ¢:

-0.62

ieu

-0.61

rots

-0.60

POSITIVE LOGITS

 cost

3.66

 costs

2.76

 Cost

2.51

cost

2.47

 costing

2.30

 Costs

2.30

Cost

2.29

 expense

1.98

 price

1.93

 expenses

1.59

Activations Density 0.020%

vocabularies associated with expenses and monetary value

references to financial expenses or costs

No Comments

No Known Activations