INDEX

Explanations

the word "implausible" and variations of it

oai_token-act-pair · gpt-3.5-turbo

terms related to implication or suggesting connections

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

Ĥ¬

-0.85

SEA

-0.79

grass

-0.77

flix

-0.77

chal

-0.72

 fixme

-0.71

etheless

-0.68

¥µ

-0.67

boarding

-0.67

Lago

-0.65

POSITIVE LOGITS

osion

1.36

oded

1.33

icating

1.21

anting

1.21

icate

1.18

icates

1.17

ausible

1.14

oding

1.14

icit

1.07

anted

1.05

Activations Density 0.014%

the word "implausible" and variations of it

terms related to implication or suggesting connections

No Comments

No Known Activations