INDEX

Explanations

terms related to historical or modern palaces

oai_token-act-pair · gpt-3.5-turbo

references to "Palace."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ELL

-0.80

rd

-0.76

went

-0.74

::::::::

-0.72

VOL

-0.71

CAST

-0.71

 Braun

-0.67

CHAT

-0.66

ivity

-0.66

rss

-0.65

POSITIVE LOGITS

 Palace

1.39

 palace

0.95

 tremend

0.80

 Hotel

0.79

odon

0.78

osaurs

0.75

 maiden

0.74

 Villa

0.73

ovo

0.72

gur

0.71

Activations Density 0.005%

terms related to historical or modern palaces

references to "Palace."

No Comments

No Known Activations