INDEX

Explanations

phrases or sentences ending with the word "t"

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "didn't" in various contexts

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

 SetTextColor

-0.71

£ı

-0.68

 Gleaming

-0.67

å£«

-0.66

soType

-0.64

 Penguin

-0.64

Reviewer

-0.63

 Reloaded

-0.62

RTX

-0.62

pter

-0.61

POSITIVE LOGITS

apest

0.98

weet

0.91

acular

0.89

cha

0.88

aper

0.87

 bother

0.87

aylor

0.86

une

0.86

unes

0.85

 necessarily

0.84

Activations Density 0.055%

phrases or sentences ending with the word "t"

instances of the phrase "didn't" in various contexts

No Comments

No Known Activations