INDEX

Explanations

mentions of locations, specifically Delhi

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Delhi."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

xon

-0.76

htaking

-0.71

chell

-0.70

ovie

-0.70

PsyNetMessage

-0.69

RH

-0.68

cies

-0.67

¶æ

-0.67

 Robertson

-0.67

 Hubbard

-0.66

POSITIVE LOGITS

 Pradesh

1.11

 Delhi

0.94

jriwal

0.89

 Punjab

0.87

 Lumpur

0.82

 crore

0.74

ascus

0.74

ki

0.72

igans

0.70

 Xuan

0.69

Activations Density 0.008%

mentions of locations, specifically Delhi

occurrences of the word "Delhi."

No Comments

No Known Activations