INDEX

Explanations

positive adjectives followed by nouns

oai_token-act-pair · gpt-3.5-turbo

expressions emphasizing the word "such" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ertodd

-0.82

kick

-0.69

ARM

-0.69

ntil

-0.68

TeX

-0.68

 Murd

-0.67

ysc

-0.64

·

-0.62

 Drum

-0.62

iliate

-0.62

POSITIVE LOGITS

ties

0.72

ities

0.70

 abundantly

0.65

 awful

0.65

 consequential

0.64

thin

0.62

 specificity

0.61

minded

0.60

vered

0.60

 sums

0.60

Activations Density 0.047%

positive adjectives followed by nouns

expressions emphasizing the word "such" in various contexts

No Comments

No Known Activations