INDEX

Explanations

terms related to size, magnitude, or extent

oai_token-act-pair · gpt-3.5-turbo

references to dimensions in various contexts

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

yah

-0.71

 scoff

-0.68

 Rouge

-0.67

ULE

-0.66

pkg

-0.65

pd

-0.63

Mist

-0.61

Lee

-0.61

ukemia

-0.60

Vul

-0.60

POSITIVE LOGITS

 dimensions

3.74

 dimension

3.61

 Dimensions

2.55

 Dimension

2.38

 dimensional

2.27

dimension

2.25

dimensional

1.74

imensional

1.65

dim

1.55

Dim

1.46

Activations Density 0.012%

terms related to size, magnitude, or extent

references to dimensions in various contexts

No Comments

No Known Activations