INDEX

Explanations

instances where someone or something is anticipating a particular outcome

oai_token-act-pair · gpt-3.5-turbo

expectation statements related to predictions or forecasts

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

papers

-0.68

odor

-0.67

otin

-0.66

crim

-0.65

ãĤ±

-0.63

Tid

-0.62

ossier

-0.62

liam

-0.62

 pseudonym

-0.61

 Awakening

-0.61

POSITIVE LOGITS

 expects

2.87

 anticip

1.08

 expect

1.03

 accepts

1.01

 expecting

0.98

expected

0.97

env

0.92

Property

0.91

 expected

0.90

 assumes

0.89

Activations Density 0.036%

instances where someone or something is anticipating a particular outcome

expectation statements related to predictions or forecasts

No Comments

No Known Activations