INDEX

Explanations

instances where something is assumed or taken as a given

oai_token-act-pair · gpt-3.5-turbo

instances of the word "assumed" in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Money

-0.75

Sche

-0.71

licks

-0.68

 deed

-0.68

alez

-0.66

Blog

-0.64

enta

-0.62

andro

-0.62

Tips

-0.62

 deeds

-0.61

POSITIVE LOGITS

 assumed

1.10

 assume

1.04

 assumes

0.86

 assuming

0.78

 "$:/

0.75

 incorrectly

0.73

 assum

0.73

llor

0.70

IELD

0.69

 assumptions

0.65

Activations Density 0.008%

instances where something is assumed or taken as a given

instances of the word "assumed" in various contexts

No Comments

No Known Activations