INDEX

Explanations

locations or places in the context

oai_token-act-pair · gpt-3.5-turbo

references to geographical locations and landmarks

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

 emoji

-0.72

 inexperienced

-0.70

 acron

-0.69

 antioxidants

-0.69

 energetic

-0.69

 disruptive

-0.68

 headline

-0.68

 reactive

-0.68

 negotiator

-0.67

 opioids

-0.67

POSITIVE LOGITS

 Island

1.40

 Castle

1.37

 Cemetery

1.36

 Cove

1.33

 Canyon

1.32

opolis

1.32

 Junction

1.30

 Harbor

1.29

 Square

1.27

 Falls

1.27

Activations Density 0.461%

locations or places in the context

references to geographical locations and landmarks

No Comments

No Known Activations