INDEX

Explanations

words related to laboratory or research environments

oai_token-act-pair · gpt-3.5-turbo

references to laboratory environments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

theless

-0.84

 Leilan

-0.75

cffffcc

-0.72

 Shades

-0.67

 Ceres

-0.67

lihood

-0.66

 Gothic

-0.66

cut

-0.63

 Marketable

-0.62

 Tart

-0.61

POSITIVE LOGITS

rador

1.39

elling

1.34

orer

1.19

yrinth

1.08

elled

1.02

rosse

1.00

orously

0.90

lab

0.88

ysis

0.88

chem

0.88

Activations Density 0.017%

words related to laboratory or research environments

references to laboratory environments

No Comments

No Known Activations