INDEX

Explanations

words related to train stations

oai_token-act-pair · gpt-3.5-turbo

references to different "stations" or transit hubs

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

 nerv

-0.66

STEM

-0.65

ortium

-0.63

 Marketable

-0.62

ojure

-0.61

URES

-0.61

usalem

-0.61

icit

-0.60

 Canaver

-0.60

erenn

-0.60

POSITIVE LOGITS

 wagon

1.11

ery

1.06

wagon

0.91

 station

0.90

 stations

0.85

ation

0.82

eering

0.82

 attendant

0.81

pole

0.80

acle

0.79

Activations Density 0.029%

words related to train stations

references to different "stations" or transit hubs

No Comments

No Known Activations