INDEX

Explanations

phrases starting with 'that' which are often used to introduce reported speech or thoughts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "that" and its associated phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

aceae

-0.71

raq

-0.70

estones

-0.67

emis

-0.67

osures

-0.67

neau

-0.66

onomic

-0.63

zman

-0.63

ggles

-0.62

ante

-0.62

POSITIVE LOGITS

"[

0.87

 whoever

0.85

'[

0.84

he

0.82

 "...

0.79

 they

0.78

 "âĢ¦

0.78

if

0.77

 sounded

0.74

nob

0.74

Activations Density 0.252%

phrases starting with 'that' which are often used to introduce reported speech or thoughts

instances of the word "that" and its associated phrases

No Comments

No Known Activations