INDEX

Explanations

political statements or arguments

oai_token-act-pair · gpt-3.5-turbo

references to significant and impactful events or conditions

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 hemor

-0.76

 zoning

-0.74

 plat

-0.70

 Borough

-0.68

 Manhattan

-0.66

 destro

-0.65

 airs

-0.65

 charm

-0.65

 playbook

-0.64

 Marketable

-0.64

POSITIVE LOGITS

ª

1.26

¹

1.25

ı

1.16

¸

1.11

³

1.09

¦

1.09

«

1.06

taboola

1.05

ł

1.04

£

1.04

Activations Density 0.087%

political statements or arguments

references to significant and impactful events or conditions

No Comments

No Known Activations