INDEX

Explanations

phrases or sentences indicating that something was influenced or motivated by something else

oai_token-act-pair · gpt-3.5-turbo

instances of the word "inspired" 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

 Canad

-0.78

alon

-0.74

 Strait

-0.74

amen

-0.70

talk

-0.67

atari

-0.66

estation

-0.65

restling

-0.64

puted

-0.63

orr

-0.61

POSITIVE LOGITS

iration

1.11

 inspiration

1.01

 inspired

0.92

 Inspired

0.84

ulously

0.83

Insp

0.82

 Cosponsors

0.81

ivated

0.76

 inspiring

0.75

urous

0.73

Activations Density 0.018%

phrases or sentences indicating that something was influenced or motivated by something else

instances of the word "inspired" in various contexts

No Comments

No Known Activations