INDEX

Explanations

instances where someone is being observed or approached by another person

oai_token-act-pair · gpt-3.5-turbo

instances of first-person narration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 redundancy

-0.75

Rai

-0.74

 defamation

-0.74

 understatement

-0.71

Origin

-0.70

 independence

-0.70

 Exile

-0.69

 Professional

-0.67

 Immunity

-0.67

 secession

-0.67

POSITIVE LOGITS

 noticed

1.41

 encountered

1.31

 spotted

1.17

 overhe

1.13

 encount

1.09

 discovered

1.07

 witnessed

1.06

 overheard

1.05

 greeted

1.02

 smelled

1.01

Activations Density 0.393%

instances where someone is being observed or approached by another person

instances of first-person narration

No Comments

No Known Activations