INDEX

Explanations

mentions of unusual or strange occurrences or behaviors

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "bizarre" and its variations in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Serv

-0.76

 Ready

-0.75

ready

-0.72

 Purch

-0.72

esp

-0.71

Pers

-0.70

uci

-0.69

agement

-0.68

serv

-0.67

Eng

-0.67

POSITIVE LOGITS

 bizarre

3.00

 ludicrous

1.87

 surreal

1.86

izarre

1.84

 outlandish

1.74

 inexplicable

1.72

 horrifying

1.72

 grotesque

1.66

 nonsensical

1.63

 absurd

1.62

Activations Density 0.036%

mentions of unusual or strange occurrences or behaviors

occurrences of the word "bizarre" and its variations in context

No Comments

No Known Activations