INDEX

Explanations

keywords related to technology products, specifically related to Amazon's Fire products

oai_token-act-pair · gpt-3.5-turbo

references to the term "Fire" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ĸļ

-0.97

ļéĨĴ

-0.78

Ń·

-0.77

atem

-0.76

 fixme

-0.72

terday

-0.71

berra

-0.71

 versa

-0.69

ansen

-0.68

aunder

-0.67

POSITIVE LOGITS

flies

1.08

fighter

0.96

 Fire

0.93

wolf

0.90

fighters

0.88

storm

0.86

 Emblem

0.85

 Explos

0.83

onica

0.82

bats

0.81

Activations Density 0.010%

keywords related to technology products, specifically related to Amazon's Fire products

references to the term "Fire" in various contexts

No Comments

No Known Activations