INDEX

Explanations

specific references to "Fed," likely referring to the Federal Reserve

oai_token-act-pair · gpt-3.5-turbo

references to the Federal Reserve

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

lihood

-0.70

lust

-0.67

 Monstrous

-0.67

 Sora

-0.66

______

-0.66

 Carbuncle

-0.66

 aliens

-0.63

Faw

-0.62

 alien

-0.62

++++++++++++++++

-0.62

POSITIVE LOGITS

erer

1.11

oras

1.02

rik

0.97

ERAL

0.96

ric

0.95

rique

0.93

eli

0.91

ricular

0.91

ered

0.88

rix

0.87

Activations Density 0.015%

specific references to "Fed," likely referring to the Federal Reserve

references to the Federal Reserve

No Comments

No Known Activations