INDEX

Explanations

phrases indicating a disclaimer or a lack of absolute certainty

oai_token-act-pair · gpt-3.5-turbo

phrases indicating conditions that are not absolute or definitive

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

banks

-0.88

busters

-0.81

met

-0.73

bern

-0.72

bah

-0.71

iry

-0.71

min

-0.70

nan

-0.69

 Hunting

-0.68

 Emin

-0.67

POSITIVE LOGITS

 necessarily

0.94

 anymore

0.80

 bothered

0.78

 detract

0.77

 involved

0.72

 conform

0.72

 guiActiveUn

0.72

omsday

0.71

 influenced

0.71

 identifiable

0.71

Activations Density 0.020%

phrases indicating a disclaimer or a lack of absolute certainty

phrases indicating conditions that are not absolute or definitive

No Comments

No Known Activations