INDEX

Explanations

interesting or noteworthy statements

oai_token-act-pair · gpt-3.5-turbo

the adverb "interestingly" and its variations, indicating a focus on presenting facts in a noteworthy manner

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 assurance

-0.87

 urgent

-0.80

 continuing

-0.77

 rush

-0.77

 assurances

-0.76

 commitment

-0.74

 âĢİ

-0.74

 steady

-0.71

 sustained

-0.70

 discipline

-0.70

POSITIVE LOGITS

 Interestingly

2.65

Interestingly

2.46

 interestingly

2.02

 Ironically

1.93

 Notably

1.78

Ironically

1.65

 Apparently

1.53

 Presumably

1.48

 Surprisingly

1.45

 Essentially

1.39

Activations Density 0.040%

interesting or noteworthy statements

the adverb "interestingly" and its variations, indicating a focus on presenting facts in a noteworthy manner

No Comments

No Known Activations