INDEX

Explanations

phrases related to implementing, arranging, or establishing things or plans

oai_token-act-pair · gpt-3.5-turbo

expressions that convey situational context or placement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

eni

-0.71

ľ

-0.69

ences

-0.68

enance

-0.64

Ł

-0.63

¼

-0.62

jc

-0.62

ents

-0.61

ĺ

-0.61

}{

-0.61

POSITIVE LOGITS

 jeopardy

1.05

 perspective

0.82

ulhu

0.72

 place

0.70

soDeliveryDate

0.69

 orbit

0.66

offensive

0.66

 seiz

0.65

 quotation

0.65

grave

0.63

Activations Density 0.094%

phrases related to implementing, arranging, or establishing things or plans

expressions that convey situational context or placement

No Comments

No Known Activations