INDEX

Explanations

words related to uncertainty or questioning

oai_token-act-pair · gpt-3.5-turbo

inquiries related to seeking information or clarity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

©¶æ

-0.76

once

-0.71

front

-0.67

mir

-0.67

}}}

-0.64

levels

-0.62

partial

-0.60

forth

-0.60

Secondly

-0.60

rocal

-0.59

POSITIVE LOGITS

 itching

0.73

INGTON

0.69

ãĥĦ

0.67

 bookmark

0.67

 downloading

0.67

ancies

0.66

itaire

0.66

 Yourself

0.64

ado

0.64

 browsing

0.64

Activations Density 0.487%

words related to uncertainty or questioning

inquiries related to seeking information or clarity

No Comments

No Known Activations