INDEX

Explanations

phrases or keywords indicating potential consequences or outcomes

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate potential consequences or outcomes

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

 Stras

-0.69

schild

-0.64

arest

-0.62

 Shal

-0.62

 Vaughn

-0.61

 tuber

-0.61

rehens

-0.61

atching

-0.60

thy

-0.59

afort

-0.59

POSITIVE LOGITS

wcs

0.86

gers

0.82

better

0.80

iments

0.74

hole

0.72

GGGG

0.71

entious

0.70

-+

0.68

ãĥĥãĥī

0.67

ging

0.65

Activations Density 0.037%

phrases or keywords indicating potential consequences or outcomes

phrases that indicate potential consequences or outcomes

No Comments

No Known Activations