INDEX

Explanations

phrases related to adding something to an existing entity

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate the action of adding something

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

 miscar

-0.74

cffffcc

-0.72

ults

-0.71

Champ

-0.67

rowth

-0.65

;;;;;;;;

-0.62

zeb

-0.62

tions

-0.62

ATIONS

-0.61

BUS

-0.61

POSITIVE LOGITS

 bolster

0.91

pload

0.83

 complement

0.82

 shore

0.76

 enhance

0.76

 compensate

0.74

 fill

0.73

 accompany

0.69

adies

0.68

 replace

0.66

Activations Density 0.142%

phrases related to adding something to an existing entity

phrases that indicate the action of adding something

No Comments

No Known Activations