INDEX

Explanations

positive emotions or appreciative expressions

oai_token-act-pair · gpt-3.5-turbo

expressions of praise, appreciation, and lamentation in the context of social and cultural commentary

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

 oath

-0.74

frog

-0.69

 violet

-0.67

 shroud

-0.66

 beads

-0.66

 ribbon

-0.65

 assassin

-0.65

 looms

-0.64

 hairs

-0.64

 oats

-0.64

POSITIVE LOGITS

ably

1.64

ational

1.32

ationally

1.27

ations

1.22

atory

1.20

ability

1.17

ating

1.16

able

1.11

antly

1.10

ately

1.10

Activations Density 0.039%

positive emotions or appreciative expressions

expressions of praise, appreciation, and lamentation in the context of social and cultural commentary

No Comments

No Known Activations