INDEX

Explanations

highlights or key aspects in a set of given text snippets

oai_token-act-pair · gpt-3.5-turbo

possessive pronouns and phrases indicating ownership or favorites

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

amation

-0.77

umption

-0.70

$$$$

-0.68

ption

-0.68

>[

-0.68

anse

-0.67

icter

-0.64

section

-0.62

tone

-0.62

iquid

-0.61

POSITIVE LOGITS

 favorite

1.20

 favourite

1.18

 favorites

1.18

 earliest

1.18

 closest

1.17

 favourites

1.12

 finest

1.09

 predecessors

1.06

 contemporaries

1.04

 greatest

1.03

Activations Density 0.127%

highlights or key aspects in a set of given text snippets

possessive pronouns and phrases indicating ownership or favorites

No Comments

No Known Activations