INDEX

Explanations

sentences expressing opinions, especially related to government policies and actions

oai_token-act-pair · gpt-3.5-turbo

statements indicating results or conclusions in a discussion or report

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 lizard

-0.69

 monop

-0.69

kid

-0.68

 plaus

-0.68

 fucked

-0.67

gou

-0.67

 captcha

-0.67

 lounge

-0.67

 pigment

-0.66

 hipp

-0.66

POSITIVE LOGITS

Speaking

1.30

Refer

1.28

 Speaking

1.20

He

1.15

Calling

1.06

 Refer

1.05

His

1.04

He

1.04

 Asked

1.04

Prime

1.03

Activations Density 0.270%

sentences expressing opinions, especially related to government policies and actions

statements indicating results or conclusions in a discussion or report

No Comments

No Known Activations