INDEX

Explanations

words related to support or endorsement

oai_token-act-pair · gpt-3.5-turbo

references to support or endorsement

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

catch

-0.70

iak

-0.67

ei

-0.64

Occ

-0.63

oj

-0.62

Osc

-0.61

acular

-0.60

species

-0.60

unc

-0.59

owl

-0.58

POSITIVE LOGITS

 backed

3.64

backed

2.16

 backing

2.10

 supported

1.94

 endorsed

1.62

 bolstered

1.46

supported

1.40

 backs

1.35

 championed

1.27

 fuelled

1.27

Activations Density 0.014%

words related to support or endorsement

references to support or endorsement

No Comments

No Known Activations