INDEX

Explanations

mentions of the word "Snow" followed by numerals representing a rating or measurement

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Snow" 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 @ 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

igious

-0.84

izabeth

-0.71

ulhu

-0.70

ï¿½ï¿½ï¿½ï¿½

-0.69

ishi

-0.68

ect

-0.66

ributed

-0.64

ista

-0.64

hetical

-0.64

ugu

-0.64

POSITIVE LOGITS

flake

1.34

 Snow

1.05

 Leopard

0.94

 Crash

0.83

dale

0.82

 Wolves

0.81

tro

0.80

storms

0.80

bats

0.79

balls

0.78

Activations Density 0.005%

mentions of the word "Snow" followed by numerals representing a rating or measurement

instances of the word "Snow" in various contexts

No Comments

No Known Activations