INDEX

Explanations

phrases related to the city of San Jose and its hockey team, the Sharks

oai_token-act-pair · gpt-3.5-turbo

mentions of "San Jose"

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

atform

-0.82

å¥³

-0.77

Els

-0.76

icles

-0.75

ãĤ¤ãĥĪ

-0.75

natureconservancy

-0.74

ischer

-0.73

ICLE

-0.68

ickle

-0.67

holder

-0.65

POSITIVE LOGITS

 Earthqu

1.07

 Sharks

0.86

Ń·

0.76

 Valley

0.73

olean

0.70

ricanes

0.69

fc

0.68

nton

0.68

mia

0.68

ppa

0.67

Activations Density 0.016%

phrases related to the city of San Jose and its hockey team, the Sharks

mentions of "San Jose"

No Comments

No Known Activations