INDEX

Explanations

questions or scenarios involving potential actions or consequences

oai_token-act-pair · gpt-3.5-turbo

interrogative contexts and queries

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

verning

-0.67

 scapego

-0.64

 controversies

-0.62

 stability

-0.61

 Gale

-0.60

 blame

-0.60

 globalization

-0.58

 substantive

-0.58

 disagreements

-0.57

 structural

-0.57

POSITIVE LOGITS

then

1.03

 THEN

0.95

 Then

0.85

Then

0.83

Repeat

0.82

Or

0.80

fax

0.80

Alternatively

0.79

Sounds

0.77

âĹ¼

0.76

Activations Density 0.969%

questions or scenarios involving potential actions or consequences

interrogative contexts and queries

No Comments

No Known Activations