INDEX

Explanations

phrases related to returning from a specific place or activity

oai_token-act-pair · gpt-3.5-turbo

instances indicating a return from a previous location or state

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

iquette

-0.82

oyer

-0.77

CHAT

-0.73

ech

-0.71

ifer

-0.68

 merits

-0.67

sylv

-0.67

 marketed

-0.65

anted

-0.64

ographic

-0.64

POSITIVE LOGITS

 whence

1.01

 vacation

0.85

 captivity

0.79

 exile

0.78

 hiatus

0.74

 overseas

0.72

 abroad

0.72

 hiber

0.71

 inspecting

0.70

 sever

0.70

Activations Density 0.081%

phrases related to returning from a specific place or activity

instances indicating a return from a previous location or state

No Comments

No Known Activations