INDEX

Explanations

terms related to geographical locations, particularly Olympia and Orion

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to locations and names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

VAT

-0.90

Naz

-0.80

 Gael

-0.77

cu

-0.75

 Manziel

-0.74

aum

-0.73

âģ

-0.72

itor

-0.72

Nig

-0.70

Ty

-0.70

POSITIVE LOGITS

 Tacoma

1.91

 Olympia

1.73

 Orion

1.72

 Paramount

1.48

 Everett

1.43

 Bernstein

1.42

 Andromeda

1.26

 Franken

1.23

romeda

1.22

 Eden

1.20

Activations Density 0.050%

terms related to geographical locations, particularly Olympia and Orion

proper nouns related to locations and names

No Comments

No Known Activations