INDEX

Explanations

phrases related to future events or plans

oai_token-act-pair · gpt-3.5-turbo

the phrase "to be" in various grammatical contexts

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

 Lines

-0.71

 Prescott

-0.70

anova

-0.65

 sqor

-0.62

typ

-0.60

pedia

-0.60

rador

-0.59

 partName

-0.58

 Stain

-0.58

 Moose

-0.58

POSITIVE LOGITS

 able

1.01

 judged

0.99

heading

0.98

fits

0.98

ige

0.97

leeve

0.97

hemoth

0.91

ijing

0.90

arers

0.90

ardless

0.88

Activations Density 0.117%

phrases related to future events or plans

the phrase "to be" in various grammatical contexts

No Comments

No Known Activations