INDEX

Explanations

comparative phrases involving the concept of "more."

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate varying degrees of comparison or intensity

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

-0.77

 Skies

-0.69

brates

-0.65

SEA

-0.63

LAN

-0.62

eming

-0.62

onde

-0.61

ãĥīãĥ©ãĤ´ãĥ³

-0.61

Ds

-0.61

heid

-0.61

POSITIVE LOGITS

 glim

0.83

 paddle

0.73

 backstory

0.72

 newsp

0.72

*/(

0.72

jriwal

0.69

 elbow

0.67

 lemon

0.67

 trickle

0.67

 overboard

0.67

Activations Density 0.159%

comparative phrases involving the concept of "more."

phrases that indicate varying degrees of comparison or intensity

No Comments

No Known Activations