INDEX

Explanations

verbs expressing uncertainty or confusion

oai_token-act-pair · gpt-3.5-turbo

expressions of confusion or distress

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

accompan

-0.72

İĭ

-0.72

arcity

-0.71

senal

-0.71

requires

-0.67

licts

-0.66

pmwiki

-0.65

eatures

-0.64

olesc

-0.63

pires

-0.63

POSITIVE LOGITS

 haha

1.21

lol

1.05

 anyways

1.02

 because

1.01

 till

1.00

:(

0.99

 afterwards

0.97

 until

0.95

because

0.93

!!

0.92

Activations Density 0.377%

verbs expressing uncertainty or confusion

expressions of confusion or distress

No Comments

No Known Activations