INDEX

Explanations

posts labeled with "Posted"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Posted" and its variations, indicating updates or entries

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

['

-0.81

 coefficients

-0.67

 internally

-0.65

 count

-0.64

 contr

-0.64

 mock

-0.63

 index

-0.62

 indexes

-0.61

 large

-0.60

 steering

-0.60

POSITIVE LOGITS

 Posted

4.22

Posted

2.04

 Logged

1.62

posted

1.38

 posted

1.24

________________

1.12

Published

1.10

 âĢİ

1.08

 Published

1.08

 Advertisement

1.06

Activations Density 0.012%

posts labeled with "Posted"

instances of the word "Posted" and its variations, indicating updates or entries

No Comments

No Known Activations