INDEX

Explanations

phrases or terms related to geographical locations, specifically regions containing directional pointers like "North" or "South."

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inventoryQuantity

-0.85

ãĤ¢ãĥ«

-0.78

TY

-0.71

ãĥİ

-0.70

Ø©

-0.69

PLIC

-0.67

TED

-0.67

ãĥĬ

-0.65

anova

-0.65

REC

-0.65

POSITIVE LOGITS

 Pole

0.98

 Indies

0.93

 Caucasus

0.93

side

0.90

 Territories

0.90

 Side

0.89

 hemisphere

0.87

 Siberian

0.87

 Coast

0.84

 Atlantic

0.84

Activations Density 0.083%

phrases or terms related to geographical locations, specifically regions containing directional pointers like "North" or "South."

No Comments

No Known Activations