INDEX

Explanations

phrases related to opinions or evaluations

oai_token-act-pair · gpt-3.5-turbo

sentences that express definitive or conclusive statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 disemb

-0.76

 nodd

-0.72

dep

-0.69

 purse

-0.67

yip

-0.67

soDeliveryDate

-0.67

 dips

-0.66

 explan

-0.66

 agent

-0.65

ogene

-0.65

POSITIVE LOGITS

 Eventually

1.14

 Ultimately

1.14

 Instead

1.11

 Thankfully

1.10

 Fortunately

1.09

 Initially

1.08

 Afterwards

1.07

 Then

1.06

 However

1.06

 Unfortunately

1.06

Activations Density 0.770%

phrases related to opinions or evaluations

sentences that express definitive or conclusive statements

No Comments

No Known Activations