INDEX

Explanations

instances of the word "overall" in the text

oai_token-act-pair · gpt-3.5-turbo

references to overall assessments or evaluations

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

hots

-0.85

rosso

-0.74

acus

-0.72

utics

-0.72

_-

-0.67

 ashore

-0.66

PAC

-0.66

pots

-0.66

bane

-0.65

friends

-0.64

POSITIVE LOGITS

 populace

1.02

 portion

0.95

 population

0.94

 extent

0.88

 element

0.88

 electorate

0.86

 equivalent

0.83

 edition

0.83

 amount

0.82

 brunt

0.81

Activations Density 0.186%

instances of the word "overall" in the text

references to overall assessments or evaluations

No Comments

No Known Activations