INDEX

Explanations

phrases and sentences expressing positive evaluation or approval

oai_token-act-pair · gpt-3.5-turbo

expressions indicating something is considered "good" or has positive qualities

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

meric

-0.69

ubi

-0.63

pora

-0.61

 Jude

-0.61

berto

-0.60

minecraft

-0.60

atars

-0.59

eni

-0.58

adium

-0.58

Peb

-0.58

POSITIVE LOGITS

illac

0.78

 indeed

0.72

 sidx

0.68

(>

0.67

sighted

0.66

 sounding

0.66

(~

0.65

 wound

0.65

 understatement

0.64

nered

0.64

Activations Density 0.099%

phrases and sentences expressing positive evaluation or approval

expressions indicating something is considered "good" or has positive qualities

No Comments

No Known Activations