INDEX

Explanations

instances of the word "quite" used in the context of the text

oai_token-act-pair · gpt-3.5-turbo

phrases indicating something that is not entirely accurate or satisfactory

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ults

-0.82

 DRAG

-0.81

selage

-0.79

EStream

-0.79

ichael

-0.78

ultimate

-0.78

adium

-0.77

ãĥĨ

-0.73

olen

-0.73

rats

-0.73

POSITIVE LOGITS

 pleasant

0.74

 spoil

0.74

icable

0.73

 enough

0.73

 bother

0.71

 bothered

0.71

spo

0.69

 neatly

0.68

 satisfactory

0.67

 anymore

0.67

Activations Density 0.007%

instances of the word "quite" used in the context of the text

phrases indicating something that is not entirely accurate or satisfactory

No Comments

No Known Activations