INDEX

Explanations

references to the country of Ethiopia

oai_token-act-pair · gpt-3.5-turbo

mentions of Ethiopia and related references

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

rooms

-0.87

iard

-0.81

pton

-0.77

naire

-0.77

busters

-0.76

ificent

-0.74

hood

-0.74

è»

-0.74

hound

-0.73

creen

-0.72

POSITIVE LOGITS

 Ethiop

0.98

 Ethiopian

0.92

idious

0.81

 Ethiopia

0.76

 Erit

0.73

herty

0.72

rican

0.68

lectic

0.68

onential

0.67

ERSON

0.66

Activations Density 0.039%

references to the country of Ethiopia

mentions of Ethiopia and related references

No Comments

No Known Activations