INDEX

Explanations

references to a specific term likely related to the name "Gonzalez" in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to Ponzi schemes

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

 Lauder

-0.74

 SPORTS

-0.71

 Borders

-0.70

ICES

-0.66

population

-0.63

OPLE

-0.61

 Skull

-0.61

 Chief

-0.61

ween

-0.60

 Primal

-0.59

POSITIVE LOGITS

alez

1.20

onz

1.02

ole

0.91

oshi

0.90

osit

0.85

onne

0.82

hee

0.82

ework

0.82

aga

0.81

kered

0.77

Activations Density 0.016%

references to a specific term likely related to the name "Gonzalez" in various contexts

references to Ponzi schemes

No Comments

No Known Activations