INDEX

Explanations

questions starting with "Was" or "were" suggesting a comparison or inquiry about past events or situations

oai_token-act-pair · gpt-3.5-turbo

questions beginning with "Was."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

otic

-0.69

entric

-0.67

enser

-0.62

 populism

-0.60

ennett

-0.59

omsday

-0.58

inas

-0.58

 resonance

-0.57

opian

-0.57

ipher

-0.57

POSITIVE LOGITS

hes

1.17

wolves

1.00

abi

0.96

 originally

0.94

hers

0.91

 formerly

0.89

wolf

0.88

 previously

0.81

ps

0.76

 born

0.75

Activations Density 0.123%

questions starting with "Was" or "were" suggesting a comparison or inquiry about past events or situations

questions beginning with "Was."

No Comments

No Known Activations