INDEX

Explanations

past tense verbs related to actions or behaviors

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.85

oldown

-0.77

osate

-0.76

kson

-0.70

adesh

-0.70

alez

-0.69

ades

-0.67

merce

-0.64

 affairs

-0.63

hower

-0.63

POSITIVE LOGITS

ered

1.22

 signs

1.09

 willingness

0.97

 glimps

0.95

 flashes

0.93

 compassion

0.91

 appreciation

0.88

 solidarity

0.87

 mercy

0.87

 reluctance

0.86

Activations Density 0.068%

past tense verbs related to actions or behaviors

No Comments

No Known Activations