INDEX

Explanations

words related to being supportive or supportiveness

oai_token-act-pair · gpt-3.5-turbo

expressions of supportiveness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

hig

-0.80

hunt

-0.74

oufl

-0.73

gob

-0.70

bole

-0.70

hun

-0.69

buck

-0.69

nit

-0.68

oqu

-0.67

hid

-0.67

POSITIVE LOGITS

 Supports

1.07

 supportive

1.04

heses

0.82

 endorsements

0.78

hetical

0.77

 enough

0.73

 SUPPORT

0.73

Support

0.73

hesis

0.73

SHIP

0.71

Activations Density 0.011%

words related to being supportive or supportiveness

expressions of supportiveness

No Comments

No Known Activations