INDEX

Explanations

positive anecdotes or sentiments

oai_token-act-pair · gpt-3.5-turbo

sentences ending in a period

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

uly

-0.69

thal

-0.68

arov

-0.67

 helper

-0.65

 correct

-0.64

apan

-0.64

ãĥķãĤ¡

-0.64

©¶æ

-0.63

 dimensional

-0.62

 defe

-0.62

POSITIVE LOGITS

 Occasionally

1.05

 Worse

1.04

 Visitors

1.01

But

0.98

 Amid

0.98

 Among

0.96

 Whereas

0.96

 Their

0.95

 Fortunately

0.95

 Those

0.94

Activations Density 0.934%

positive anecdotes or sentiments

sentences ending in a period

No Comments

No Known Activations