INDEX

Explanations

words related to documentation or recording of information

oai_token-act-pair · gpt-3.5-turbo

the term "documented" 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

 Flavoring

-0.77

next

-0.73

iance

-0.67

anta

-0.65

soft

-0.64

itiveness

-0.64

rade

-0.63

 Reply

-0.61

antry

-0.61

instead

-0.61

POSITIVE LOGITS

 documented

3.49

documented

2.38

 documenting

1.85

 recorded

1.76

 publicized

1.54

 documentation

1.45

 reported

1.44

 demonstrated

1.44

 verified

1.43

 researched

1.39

Activations Density 0.012%

words related to documentation or recording of information

the term "documented" in various contexts

No Comments

No Known Activations