INDEX

Explanations

published dates

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Published" along with associated publication dates

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

ixel

-0.77

utic

-0.76

ovable

-0.75

porary

-0.72

mini

-0.72

oles

-0.71

oso

-0.70

hov

-0.70

arden

-0.69

antics

-0.69

POSITIVE LOGITS

âĸ¬âĸ¬

0.78

âĸ¬

0.76

à¨

0.74

NESS

0.71

 Date

0.70

 aloud

0.70

lishing

0.69

à©

0.68

rul

0.67

Published

0.67

Activations Density 0.019%

published dates

occurrences of the word "Published" along with associated publication dates

No Comments

No Known Activations