INDEX

Explanations

occurrences of the word "we" with a high level of activation

oai_token-act-pair · gpt-3.5-turbo

instances of the pronoun "we."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 gratification

-0.80

quo

-0.71

 anonymity

-0.68

 citation

-0.67

 pedoph

-0.65

 citations

-0.64

 contradictions

-0.63

LSD

-0.63

 conflicts

-0.61

 sucker

-0.61

POSITIVE LOGITS

bsite

1.49

eping

1.18

aving

1.17

ldon

1.16

lder

1.14

igh

1.13

akening

1.11

aning

1.10

eks

1.06

ighed

1.05

Activations Density 0.098%

occurrences of the word "we" with a high level of activation

instances of the pronoun "we."

No Comments

No Known Activations