INDEX

Explanations

phrases related to prizes or rewards

oai_token-act-pair · gpt-3.5-turbo

mentions of various prizes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Sakuya

-0.73

aturdays

-0.73

 Ange

-0.72

icago

-0.70

66666666

-0.69

iants

-0.69

den

-0.69

rica

-0.67

nian

-0.65

onica

-0.64

POSITIVE LOGITS

 awarded

1.09

 prize

1.05

 prizes

1.04

 winner

0.94

 winners

0.93

 laureate

0.91

 Winner

0.91

 Winners

0.90

 purse

0.86

 awards

0.85

Activations Density 0.021%

phrases related to prizes or rewards

mentions of various prizes

No Comments

No Known Activations