INDEX

Explanations

instances where the word "before" is mentioned

oai_token-act-pair · gpt-3.5-turbo

the word "before" 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

ocated

-0.56

ocations

-0.56

greg

-0.55

nic

-0.55

emonic

-0.55

eland

-0.54

purpose

-0.53

Limited

-0.53

ocal

-0.52

Constructed

-0.52

POSITIVE LOGITS

 before

2.98

before

2.54

 BEFORE

2.32

 beforehand

1.96

 Before

1.84

 prior

1.81

Before

1.67

 preceding

1.32

 preceded

1.26

 afterwards

1.25

Activations Density 0.063%

instances where the word "before" is mentioned

the word "before" in various contexts

No Comments

No Known Activations