INDEX

Explanations

phrases related to technology and specifically mentions of Google and its various services

oai_token-act-pair · gpt-3.5-turbo

references to Google and its executives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ollah

-0.89

rentices

-0.87

 Hezbollah

-0.85

bilt

-0.83

ewitness

-0.81

abol

-0.76

chenko

-0.75

zbollah

-0.75

merga

-0.75

 Hassan

-0.75

POSITIVE LOGITS

 Google

2.12

Google

2.05

 Gmail

1.82

 Alphabet

1.80

 Chromebook

1.60

google

1.57

 Chrome

1.53

 google

1.53

 Android

1.50

phabet

1.41

Activations Density 0.260%

phrases related to technology and specifically mentions of Google and its various services

references to Google and its executives

No Comments

No Known Activations