INDEX

Explanations

superlatives denoting the highest degree or amount of something

oai_token-act-pair · gpt-3.5-turbo

phrases that quantify or emphasize the concept of "most" 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

heid

-0.84

rompt

-0.81

amily

-0.77

cia

-0.74

pload

-0.74

raine

-0.70

oak

-0.69

arent

-0.68

kas

-0.67

arity

-0.67

POSITIVE LOGITS

 bang

0.85

 expensive

0.84

 egregious

0.83

 probable

0.81

 likely

0.81

 stringent

0.80

 rigorous

0.78

 efficient

0.76

 comprehensive

0.76

 exciting

0.76

Activations Density 0.047%

superlatives denoting the highest degree or amount of something

phrases that quantify or emphasize the concept of "most" in various contexts

No Comments

No Known Activations