INDEX

Explanations

phrases expressing occasional occurrences or variability

oai_token-act-pair · gpt-3.5-turbo

punctuations and conjunctions that indicate complexities in narratives

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

 sqor

-0.60

derived

-0.58

atable

-0.56

atform

-0.55

arden

-0.55

Fund

-0.54

 citiz

-0.54

zens

-0.54

elfare

-0.53

achelor

-0.53

POSITIVE LOGITS

sometimes

1.94

 Sometimes

1.93

 sometimes

1.92

Sometimes

1.83

 Occasionally

1.63

usually

1.54

 occasionally

1.48

 Often

1.45

often

1.45

Often

1.43

Activations Density 0.377%

phrases expressing occasional occurrences or variability

punctuations and conjunctions that indicate complexities in narratives

No Comments

No Known Activations