INDEX

Explanations

words related to people, such as verbs like 'are' and 'were'

oai_token-act-pair · gpt-3.5-turbo

the verb "to be" in different forms and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pedia

-0.75

imation

-0.70

ricks

-0.68

Geh

-0.63

ooters

-0.62

Footnote

-0.62

 goodbye

-0.62

fail

-0.62

 cancellation

-0.61

wake

-0.60

POSITIVE LOGITS

tein

0.82

 fluent

0.73

dinand

0.72

 supposed

0.70

held

0.67

 subscribed

0.67

 behold

0.67

 married

0.67

 caught

0.66

 blinded

0.66

Activations Density 0.186%

words related to people, such as verbs like 'are' and 'were'

the verb "to be" in different forms and contexts

No Comments

No Known Activations