INDEX

Explanations

phrases related to specific names or entities

oai_token-act-pair · gpt-3.5-turbo

sentences that end with a period, indicating statement conclusions

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

 baskets

-0.81

 affordability

-0.76

 attrition

-0.75

 rentals

-0.73

 stocking

-0.73

 allowance

-0.73

 prelim

-0.72

 prisons

-0.72

 connectivity

-0.72

 restraints

-0.71

POSITIVE LOGITS

His

1.65

He

1.54

Born

1.39

 Born

1.36

 Known

1.32

 Himself

1.28

 Surely

1.28

 Sadly

1.25

 Though

1.25

His

1.24

Activations Density 0.317%

phrases related to specific names or entities

sentences that end with a period, indicating statement conclusions

No Comments

No Known Activations