INDEX

Explanations

phrases related to constraints, limits, and restrictions

oai_token-act-pair · gpt-3.5-turbo

concepts related to limitations or boundaries

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 à¤

-0.92

 Ãī

-0.79

 ..........

-0.68

 Dhabi

-0.66

 Krishna

-0.65

da

-0.63

heet

-0.62

INGTON

-0.62

leaders

-0.61

stown

-0.60

POSITIVE LOGITS

 allowable

1.14

 permissible

1.06

 scope

1.00

 exclus

0.86

 allowed

0.85

 maximum

0.85

 permitted

0.84

 limits

0.83

 exceptions

0.83

 freedoms

0.82

Activations Density 0.221%

phrases related to constraints, limits, and restrictions

concepts related to limitations or boundaries

No Comments

No Known Activations