INDEX

Explanations

proper nouns, specifically the name "Sean"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Sean."

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

inarily

-0.82

drm

-0.74

yrinth

-0.74

icted

-0.72

hov

-0.70

 subp

-0.69

 circ

-0.68

 tender

-0.68

 pathological

-0.65

VALUE

-0.64

POSITIVE LOGITS

 Hannity

1.05

apon

1.00

Sean

0.96

 Spicer

0.89

gan

0.77

eman

0.74

 Patrick

0.73

lihood

0.73

 Hayes

0.72

 Steele

0.72

Activations Density 0.013%

proper nouns, specifically the name "Sean"

occurrences of the name "Sean."

No Comments

No Known Activations