INDEX

Explanations

names of specific geographical locations

oai_token-act-pair · gpt-3.5-turbo

names and terms associated with specific historical figures and locations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

uality

-0.82

istant

-0.81

hematically

-0.80

Ã¨re

-0.79

hered

-0.77

wo

-0.77

arians

-0.75

agen

-0.75

ory

-0.74

ancy

-0.74

POSITIVE LOGITS

vernment

0.90

 Lumpur

0.83

emouth

0.82

 Weasley

0.80

bably

0.75

bom

0.73

 bang

0.70

 redes

0.70

 boom

0.67

 leaflets

0.67

Activations Density 0.019%

names of specific geographical locations

names and terms associated with specific historical figures and locations

No Comments

No Known Activations