INDEX

Explanations

comparisons and descriptions using the word "like"

oai_token-act-pair · gpt-3.5-turbo

phrases that describe experiences or perceptions of "what it is like" to be in various situations

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

arta

-0.73

Bot

-0.70

ondo

-0.63

 doubt

-0.62

DK

-0.61

Mat

-0.60

ho

-0.60

 Plain

-0.59

 answ

-0.58

 dispute

-0.58

POSITIVE LOGITS

liest

0.70

 unfolding

0.70

antes

0.66

 Melania

0.62

âĹ¼

0.62

 thro

0.61

abouts

0.60

 coloring

0.59

 culturally

0.59

 encountering

0.58

Activations Density 0.063%

comparisons and descriptions using the word "like"

phrases that describe experiences or perceptions of "what it is like" to be in various situations

No Comments

No Known Activations