INDEX

Explanations

quotes or direct speech punctuation

oai_token-act-pair · gpt-3.5-turbo

quoted speech or dialogue in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 favor

-0.96

 honors

-0.85

 undet

-0.82

 eleph

-0.82

 favored

-0.81

 bonded

-0.80

 footing

-0.80

 exper

-0.79

 unsus

-0.79

 satell

-0.79

POSITIVE LOGITS

But

1.73

However

1.72

They

1.72

We

1.71

Obviously

1.69

It

1.69

Therefore

1.68

Hopefully

1.67

There

1.66

Certainly

1.65

Activations Density 0.116%

quotes or direct speech punctuation

quoted speech or dialogue in the text

No Comments

No Known Activations