INDEX

Explanations

divisions and categorizations in text

oai_token-act-pair · gpt-3.5-turbo

references to divisions and classifications within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 resurgence

-0.61

 Freak

-0.61

ItemTracker

-0.60

odore

-0.60

 understatement

-0.60

quickShipAvailable

-0.59

 fumes

-0.58

THING

-0.57

Blade

-0.57

abuse

-0.56

POSITIVE LOGITS

 ranging

1.22

 varying

1.16

 separated

1.14

 spaced

1.14

 consisting

1.10

 arranged

1.08

 differing

1.07

 depending

1.05

 apiece

1.02

 spanning

1.01

Activations Density 0.298%

divisions and categorizations in text

references to divisions and classifications within various contexts

No Comments

No Known Activations