INDEX

Explanations

phrases related to time durations

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "over" followed by a numerical value

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

":[

-0.73

 Lauder

-0.72

arium

-0.68

illary

-0.66

Doe

-0.65

Express

-0.64

 Solution

-0.61

Â¯Â¯Â¯Â¯

-0.61

 lured

-0.61

UES

-0.59

POSITIVE LOGITS

drive

1.11

hang

1.09

priced

1.07

valued

0.98

whelming

0.97

loading

0.96

tones

0.94

capacity

0.94

haul

0.90

grown

0.90

Activations Density 0.068%

phrases related to time durations

phrases that include the word "over" followed by a numerical value

No Comments

No Known Activations