INDEX

Explanations

mentions of specific names or authors

oai_token-act-pair · gpt-3.5-turbo

proper names and references to authors or contributors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

soDeliveryDate

-0.83

 tooth

-0.75

 undermin

-0.74

 demoral

-0.72

 equivalent

-0.67

 steroid

-0.67

 obstruction

-0.66

 shooters

-0.66

 extremists

-0.66

 descendants

-0.65

POSITIVE LOGITS

ipel

0.91

(@

0.89

inus

0.82

 Photography

0.82

iatus

0.80

nw

0.80

ertodd

0.78

HCR

0.74

cerpt

0.73

 Originally

0.73

Activations Density 0.414%

mentions of specific names or authors

proper names and references to authors or contributors

No Comments

No Known Activations