INDEX

Explanations

mention of specific events or details in news articles

oai_token-act-pair · gpt-3.5-turbo

references to leadership positions and transitions

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

."[

-0.68

+.

-0.64

".[

-0.63

%.

-0.63

.ï¿½

-0.62

usercontent

-0.61

".

-0.60

.).

-0.60

.<

-0.59

!".

-0.59

POSITIVE LOGITS

wealth

0.52

leaf

0.50

pires

0.49

bernatorial

0.48

urances

0.48

 sequ

0.47

 wedding

0.46

Ĭ±

0.46

©¶æ

0.46

iph

0.45

Activations Density 2.446%

mention of specific events or details in news articles

references to leadership positions and transitions

No Comments

No Known Activations