INDEX

Explanations

phrases related to expressing different viewpoints or arguments

oai_token-act-pair · gpt-3.5-turbo

instances of uncertainty or caution in statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

STATE

-0.74

roup

-0.67

=#

-0.60

 guiActiveUnfocused

-0.60

MpServer

-0.60

awa

-0.59

ews

-0.59

bris

-0.59

../

-0.58

fty

-0.58

POSITIVE LOGITS

 however

1.54

 moreover

1.38

 alas

1.31

 meanwhile

1.23

 incidentally

1.21

 though

1.13

 therefore

1.07

 unsurprisingly

1.06

huh

1.04

 according

1.04

Activations Density 0.388%

phrases related to expressing different viewpoints or arguments

instances of uncertainty or caution in statements

No Comments

No Known Activations