INDEX

Explanations

words related to "Hat"

oai_token-act-pair · gpt-3.5-turbo

multiple occurrences of the word "Hat" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Interstitial

-0.80

ucha

-0.79

CRIP

-0.75

ngth

-0.72

 proposition

-0.67

 recharge

-0.66

 cell

-0.65

 electroly

-0.64

 comed

-0.63

 solvent

-0.61

POSITIVE LOGITS

Hat

3.97

Hat

2.93

hat

2.14

hat

2.07

 Hats

1.95

 hats

1.64

 Mask

1.22

 Helmet

1.17

 Coat

1.17

 Shirt

1.16

Activations Density 0.017%

words related to "Hat"

multiple occurrences of the word "Hat" in various contexts

No Comments

No Known Activations