INDEX

Explanations

phrases indicating failure or disappointing outcomes

oai_token-act-pair · gpt-3.5-turbo

negations or expressions of something not being done or absent

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

éĥ

-0.68

æĥ

-0.67

rift

-0.65

Pierre

-0.65

itor

-0.63

CRIP

-0.63

++++

-0.62

Rap

-0.61

 weap

-0.60

 Spectrum

-0.60

POSITIVE LOGITS

epad

1.11

icably

1.08

yet

1.08

icable

1.07

 been

1.01

yet

0.96

 necessarily

0.96

hin

0.90

 gotten

0.85

 fared

0.83

Activations Density 0.055%

phrases indicating failure or disappointing outcomes

negations or expressions of something not being done or absent

No Comments

No Known Activations