INDEX

Explanations

terms related to official inspections or assessments

oai_token-act-pair · gpt-3.5-turbo

mentions of inspections

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

uscript

-0.78

thood

-0.73

peak

-0.70

erry

-0.69

corn

-0.68

adding

-0.66

 Crimson

-0.66

idal

-0.65

---------

-0.65

pop

-0.64

POSITIVE LOGITS

 inspection

1.10

 inspections

1.00

 inspecting

0.86

 inspected

0.86

 inspect

0.83

 Inspection

0.83

 screenings

0.78

rador

0.78

 technician

0.78

 exam

0.77

Activations Density 0.024%

terms related to official inspections or assessments

mentions of inspections

No Comments

No Known Activations