INDEX

Explanations

URLs related to Google's various services

oai_token-act-pair · gpt-3.5-turbo

instances of the word "google" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

stood

-0.85

ais

-0.78

rone

-0.74

 Principle

-0.66

 Ninth

-0.64

earned

-0.63

lined

-0.62

arist

-0.62

Sox

-0.61

due

-0.61

POSITIVE LOGITS

images

0.72

usercontent

0.72

¿½

0.70

OGR

0.68

ãĥ£

0.68

earth

0.68

gging

0.67

ijn

0.67

²¾

0.67

ificate

0.66

Activations Density 0.011%

URLs related to Google's various services

instances of the word "google" across various contexts

No Comments

No Known Activations