INDEX

Explanations

news about various incidents or events happening in different locations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the occurrence of incidents or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ĪĴ

-0.71

 Required

-0.64

ativity

-0.60

ħĭ

-0.59

peat

-0.59

hee

-0.59

quant

-0.58

iago

-0.58

reat

-0.57

hetically

-0.57

POSITIVE LOGITS

 shortly

1.17

 midday

1.15

 Wednesday

1.04

 Saturday

1.04

 Tuesday

1.04

Aug

1.03

 Sept

1.03

 August

1.01

 Thursday

1.01

 Monday

1.01

Activations Density 0.136%

news about various incidents or events happening in different locations

phrases indicating the occurrence of incidents or events

No Comments

No Known Activations