INDEX

Explanations

locations or places, particularly involving transportation hubs like stations

oai_token-act-pair · gpt-3.5-turbo

mentions of various stations and their associated contexts

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

pace

-0.99

ãĥ£

-0.89

onymous

-0.77

umers

-0.75

umen

-0.71

ussen

-0.70

assic

-0.69

acies

-0.69

risome

-0.69

ãĥ¥

-0.68

POSITIVE LOGITS

rador

0.86

swick

0.81

 Sauce

0.74

Hunt

0.72

 Sapphire

0.71

Keys

0.64

 Flavoring

0.64

eers

0.62

SQU

0.62

 Yard

0.62

Activations Density 0.043%

locations or places, particularly involving transportation hubs like stations

mentions of various stations and their associated contexts

No Comments

No Known Activations