INDEX

Explanations

the preposition "to"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "to"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 vulner

-0.70

 Niet

-0.67

 Seym

-0.65

 millenn

-0.61

 mortg

-0.59

Gaw

-0.59

 disadvant

-0.58

 Azerb

-0.56

 prest

-0.55

 campaigned

-0.55

POSITIVE LOGITS

ggles

1.00

pload

0.89

ilet

0.86

wered

0.86

 enable

0.81

pless

0.79

 RELEASE

0.79

ppers

0.78

 provide

0.77

 learn

0.77

Activations Density 0.041%

the preposition "to"

occurrences of the word "to"

No Comments

No Known Activations