INDEX

Explanations

words related to physical locations or rooms

oai_token-act-pair · gpt-3.5-turbo

references to physical locations or spaces

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

awed

-0.72

 reconnect

-0.68

 Kickstarter

-0.64

 kindred

-0.62

 punishing

-0.62

 devastating

-0.62

haps

-0.61

 dime

-0.61

 compelling

-0.61

 frivolous

-0.60

POSITIVE LOGITS

room

1.34

rooms

1.09

idge

0.91

stairs

0.90

bell

0.82

 upstairs

0.77

house

0.77

 Room

0.76

coe

0.76

Room

0.76

Activations Density 0.008%

words related to physical locations or rooms

references to physical locations or spaces

No Comments

No Known Activations