INDEX

Explanations

proper nouns representing people

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals or characters in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

termin

-0.74

pad

-0.68

trap

-0.66

âĵĺ

-0.65

stage

-0.64

 inhibitor

-0.62

MSN

-0.62

rend

-0.61

FN

-0.60

 pend

-0.60

POSITIVE LOGITS

soever

2.19

 thou

0.73

we

0.73

he

0.70

 Hague

0.68

izens

0.67

 journalists

0.66

 observers

0.65

 relatives

0.65

 Chomsky

0.65

Activations Density 0.035%

proper nouns representing people

references to specific individuals or characters in the text

No Comments

No Known Activations