INDEX

Explanations

phrases related to information access and specific websites or URLs

oai_token-act-pair · gpt-3.5-turbo

references to access or control of information

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

ashtra

-0.68

ufact

-0.67

 brim

-0.66

pton

-0.65

 ivory

-0.64

emouth

-0.62

abal

-0.62

ulhu

-0.61

zona

-0.60

abiding

-0.60

POSITIVE LOGITS

Ö¼

0.68

illes

0.68

earances

0.67

 Agency

0.62

Charges

0.62

åħī

0.62

 BASE

0.61

ources

0.60

altern

0.59

ById

0.58

Activations Density 0.349%

phrases related to information access and specific websites or URLs

references to access or control of information

No Comments

No Known Activations