INDEX

Explanations

mentions of dolphins

oai_token-act-pair · gpt-3.5-turbo

references to dolphins

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

 Flavoring

-0.83

Interstitial

-0.83

eger

-0.74

ablishment

-0.74

OHN

-0.71

chen

-0.71

iah

-0.69

haar

-0.68

ISTER

-0.67

ures

-0.67

POSITIVE LOGITS

 dolphin

1.39

 dolphins

1.23

 whale

1.04

iform

0.94

arium

0.82

 whales

0.82

 Dolphin

0.81

olphin

0.79

 Swim

0.77

 tuna

0.75

Activations Density 0.011%

mentions of dolphins

references to dolphins

No Comments

No Known Activations