INDEX

Explanations

phrases related to quantification or comparison

oai_token-act-pair · gpt-3.5-turbo

expressions of quantity and inclusivity

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

edin

-0.78

habi

-0.74

cpp

-0.72

azeera

-0.70

aunder

-0.69

zag

-0.68

 sidx

-0.66

utterstock

-0.66

igslist

-0.65

ractor

-0.64

POSITIVE LOGITS

)))

0.71

 writers

0.70

 astronomers

0.67

 disgruntled

0.67

 scientists

0.66

 disparate

0.65

 commentators

0.64

 technicians

0.64

ONE

0.63

 programmers

0.63

Activations Density 0.378%

phrases related to quantification or comparison

expressions of quantity and inclusivity

No Comments

No Known Activations