INDEX

Explanations

apologies and expressions of regret

oai_token-act-pair · gpt-3.5-turbo

expressions of apology and regret

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

\">

-0.78

 guiActiveUn

-0.76

rones

-0.75

 sightings

-0.74

qi

-0.73

 Farming

-0.71

 mosqu

-0.71

ancies

-0.71

ndum

-0.71

population

-0.71

POSITIVE LOGITS

 apologized

2.08

 apology

2.07

 apologize

1.95

 apologizing

1.92

 apologise

1.91

 apologies

1.91

 apologised

1.91

 remorse

1.80

 regrets

1.69

 regret

1.67

Activations Density 0.671%

apologies and expressions of regret

expressions of apology and regret

No Comments

No Known Activations