INDEX

Explanations

personal pronouns

oai_token-act-pair · gpt-3.5-turbo

expressions of personal opinion or belief

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

VERTISEMENT

-0.67

 Deadline

-0.60

 gearing

-0.58

ÃįÃį

-0.56

inventoryQuantity

-0.55

 resignation

-0.53

WAR

-0.52

Pwr

-0.52

 entreprene

-0.51

 goodbye

-0.51

POSITIVE LOGITS

 guessed

1.01

 confuse

0.90

 misinterpret

0.90

 interpret

0.86

 know

0.84

 interpreted

0.83

 invented

0.83

 misunderstand

0.82

 equate

0.82

 pronounce

0.81

Activations Density 0.500%

personal pronouns

expressions of personal opinion or belief

No Comments

No Known Activations