INDEX

Explanations

terms related to conducting inspections

oai_token-act-pair · gpt-3.5-turbo

terms related to inspections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

peak

-0.83

corn

-0.71

uscript

-0.70

thood

-0.70

---------

-0.68

ctory

-0.68

 Crimson

-0.67

frey

-0.66

yip

-0.66

birth

-0.65

POSITIVE LOGITS

 inspection

1.23

 inspections

1.04

 inspected

1.03

 inspecting

1.01

 Inspection

0.97

 inspect

0.95

 Inspect

0.87

 inspector

0.85

eatures

0.76

ourgeois

0.75

Activations Density 0.012%

terms related to conducting inspections

terms related to inspections

No Comments

No Known Activations