INDEX

Explanations

technology-related terms, specifically focusing on actions taken by Microsoft

oai_token-act-pair · gpt-3.5-turbo

mentions of Microsoft and its product-related actions or announcements

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

fur

-0.76

geries

-0.65

scar

-0.65

omore

-0.63

 Kurd

-0.63

nat

-0.62

inducing

-0.61

ikh

-0.61

 scar

-0.60

ysis

-0.60

POSITIVE LOGITS

 Gleaming

0.82

 unveil

0.82

 advertise

0.79

 manufactures

0.77

Ħ¢

0.77

 partnerships

0.74

 antitrust

0.72

NYSE

0.72

ãĤ´ãĥ³

0.71

DAQ

0.70

Activations Density 0.467%

technology-related terms, specifically focusing on actions taken by Microsoft

mentions of Microsoft and its product-related actions or announcements

No Comments

No Known Activations