INDEX

Explanations

occurrences of the word "parked" in various contexts related to vehicles or locations

oai_token-act-pair · gpt-3.5-turbo

references to vehicles and the action of parking

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pter

-0.72

 diagn

-0.65

 Neck

-0.64

 educators

-0.61

 Homs

-0.61

 continuum

-0.61

 Scha

-0.61

Hol

-0.61

 Thur

-0.61

onom

-0.61

POSITIVE LOGITS

 repay

1.87

 repaid

1.71

 parked

1.61

 lodged

1.09

 unse

1.04

 unlocked

1.01

 crowdfunding

0.98

 repayment

0.98

 sadd

0.96

oco

0.92

Activations Density 0.043%

occurrences of the word "parked" in various contexts related to vehicles or locations

references to vehicles and the action of parking

No Comments

No Known Activations