INDEX

Explanations

references to large quantities or groups of people

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "thousands" as an indicator of large groups of people or significant numbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

-0.67

 Redemption

-0.66

 hegemony

-0.66

 symmetry

-0.66

agate

-0.63

iban

-0.62

 Sense

-0.61

agame

-0.61

mentioned

-0.61

shit

-0.61

POSITIVE LOGITS

 ILCS

0.83

 kilometers

0.83

 thousands

0.83

 miles

0.82

ourcing

0.80

 kilometres

0.79

 hectares

0.78

ourced

0.77

 pounds

0.74

 lashes

0.73

Activations Density 0.025%

references to large quantities or groups of people

occurrences of the word "thousands" as an indicator of large groups of people or significant numbers

No Comments

No Known Activations