INDEX

Explanations

the pronoun "it" when mentioned with varying levels of importance

oai_token-act-pair · gpt-3.5-turbo

references to a specific subject or topic that is being discussed

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

veyard

-0.61

mobile

-0.60

Band

-0.58

pta

-0.58

idth

-0.58

rams

-0.58

 Passenger

-0.58

ready

-0.57

Car

-0.57

 Herrera

-0.56

POSITIVE LOGITS

alian

1.10

chy

1.03

self

1.02

ueller

0.93

unes

0.90

atic

0.87

atical

0.80

displayText

0.78

 happening

0.77

 tremend

0.77

Activations Density 0.180%

the pronoun "it" when mentioned with varying levels of importance

references to a specific subject or topic that is being discussed

No Comments

No Known Activations