INDEX

Explanations

words related to potential outcomes or consequences

oai_token-act-pair · gpt-3.5-turbo

conditional phrases expressing potential outcomes or possibilities

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

Maker

-0.71

raining

-0.66

 Yards

-0.64

core

-0.63

orno

-0.61

 Made

-0.61

washer

-0.61

ving

-0.61

got

-0.60

 honors

-0.60

POSITIVE LOGITS

 feas

1.16

 conce

1.10

berra

1.04

adian

1.01

 conclud

0.95

 tremend

0.93

be

0.92

 potentially

0.92

 foresee

0.91

 hypot

0.90

Activations Density 0.082%

words related to potential outcomes or consequences

conditional phrases expressing potential outcomes or possibilities

No Comments

No Known Activations