INDEX

Explanations

phrases indicating a range of possibilities or outcomes based on different conditions

oai_token-act-pair · gpt-3.5-turbo

conditional phrases or statements about various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

eva

-0.76

kee

-0.70

ãĥĨ

-0.69

 Advertisement

-0.68

ãĥį

-0.67

enko

-0.66

 âĵĺ

-0.66

ãĥ¯

-0.64

lead

-0.64

FT

-0.63

POSITIVE LOGITS

 circumstance

1.22

 circumstances

1.16

 severity

1.03

how

1.00

 context

0.98

 whether

0.97

 availability

0.96

 factors

0.91

 geography

0.85

 configuration

0.85

Activations Density 0.114%

phrases indicating a range of possibilities or outcomes based on different conditions

conditional phrases or statements about various subjects

No Comments

No Known Activations