INDEX

Explanations

verbs or phrases indicating a desire or intention to take a specific action

oai_token-act-pair · gpt-3.5-turbo

phrases that express desires or intentions to perform actions

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

soDeliveryDate

-0.71

banks

-0.70

grounds

-0.70

quickShipAvailable

-0.70

front

-0.68

wikipedia

-0.67

ãĥ³ãĤ¸

-0.66

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.64

written

-0.64

iously

-0.64

POSITIVE LOGITS

 browse

1.02

 keep

0.98

 maximize

0.94

 download

0.94

 qualify

0.93

 subscribe

0.93

 give

0.91

 ride

0.90

 make

0.89

 indulge

0.89

Activations Density 0.115%

verbs or phrases indicating a desire or intention to take a specific action

phrases that express desires or intentions to perform actions

No Comments

No Known Activations