INDEX

Explanations

phrases related to expressing personal opinion or position

oai_token-act-pair · gpt-3.5-turbo

repetitive phrases starting with the word "a."

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

 Jagu

-0.89

Els

-0.82

Hamilton

-0.79

Ingredients

-0.76

metadata

-0.75

andon

-0.75

mares

-0.72

bots

-0.71

ðŁĺ

-0.69

ORED

-0.69

POSITIVE LOGITS

 result

1.04

 reminder

1.01

 standalone

0.95

 member

0.89

 consequence

0.87

 footnote

0.86

 disclaimer

0.86

 testament

0.83

 viewer

0.82

 workaround

0.80

Activations Density 0.066%

phrases related to expressing personal opinion or position

repetitive phrases starting with the word "a."

No Comments

No Known Activations