INDEX

Explanations

dates and time-related information

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate significant events or changes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abus

-0.46

latable

-0.44

ruption

-0.43

gins

-0.43

ggle

-0.42

 Helmet

-0.42

ilege

-0.42

ergy

-0.42

ilet

-0.41

Chinese

-0.41

POSITIVE LOGITS

ãĥĩãĤ£

0.60

 nonetheless

0.51

pse

0.48

ãĥĺ

0.48

ãĥĦ

0.47

srf

0.47

 compr

0.45

bably

0.45

pal

0.44

conservancy

0.43

Activations Density 5.098%

dates and time-related information

phrases that indicate significant events or changes

No Comments

No Known Activations