INDEX

Explanations

references to a specific location, which is Munich

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Munich"

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

BOOK

-0.99

ordinate

-0.88

::::::::

-0.82

uably

-0.76

onies

-0.73

liest

-0.72

avery

-0.71

~~~~~~~~~~~~~~~~

-0.70

served

-0.68

cause

-0.66

POSITIVE LOGITS

 Munich

1.31

 MÃ¼

0.93

 Lumpur

0.90

sburg

0.88

 Dortmund

0.76

£ı

0.73

 Vienna

0.69

 Calling

0.69

 Dresden

0.68

stadt

0.68

Activations Density 0.007%

references to a specific location, which is Munich

occurrences of the word "Munich"

No Comments

No Known Activations