INDEX

Explanations

numerical figures mentioned in texts

oai_token-act-pair · gpt-3.5-turbo

quantitative data related to large groups of people or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

inventoryQuantity

-0.68

viz

-0.61

 probabilities

-0.55

 appre

-0.54

nep

-0.53

 defe

-0.52

vom

-0.51

 conve

-0.51

 abbre

-0.50

 Drops

-0.50

POSITIVE LOGITS

 million

1.19

million

1.18

1.14

 lakh

1.09

1.06

1.05

1.01

0.99

Activations Density 0.098%

numerical figures mentioned in texts

quantitative data related to large groups of people or statistics

No Comments

No Known Activations