INDEX

Explanations

instances where a word is followed by 'to'

oai_token-act-pair · gpt-3.5-turbo

recurring phrases or structures that signify direction or purpose

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pict

-0.68

pie

-0.67

Got

-0.66

soDeliveryDate

-0.65

reports

-0.65

need

-0.65

furt

-0.63

typ

-0.62

 packets

-0.61

 stru

-0.60

POSITIVE LOGITS

ggles

1.18

 ensure

1.12

 maximize

1.12

pload

1.10

 celebrate

1.09

 accommodate

1.07

 promote

1.04

adies

1.04

 compensate

1.04

 minimize

1.03

Activations Density 0.880%

instances where a word is followed by 'to'

recurring phrases or structures that signify direction or purpose

No Comments

No Known Activations