INDEX

Explanations

phrases indicating personal opinions or beliefs

oai_token-act-pair · gpt-3.5-turbo

the repeated phrase "I think that."

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

hips

-0.71

istance

-0.68

ãĥ©ãĥ³

-0.64

EMBER

-0.63

Guard

-0.62

ãĥīãĥ©

-0.61

aq

-0.60

 Directions

-0.60

orah

-0.59

IELD

-0.59

POSITIVE LOGITS

cher

0.79

 justifies

0.77

 contradicts

0.73

izoph

0.73

 translates

0.70

 applies

0.66

 sounds

0.66

 settles

0.66

 sounded

0.64

 undermines

0.63

Activations Density 0.259%

phrases indicating personal opinions or beliefs

the repeated phrase "I think that."

No Comments

No Known Activations