INDEX

Explanations

instances of years or timeframes

oai_token-act-pair · gpt-3.5-turbo

references to time indicators such as "year" and "month."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

esse

-0.75

 anecd

-0.74

gotten

-0.71

-0.69

 similarities

-0.67

had

-0.60

 recently

-0.59

 superf

-0.58

Dys

-0.57

 similarity

-0.57

POSITIVE LOGITS

 onwards

0.74

study

0.73

mie

0.72

sale

0.71

mitt

0.70

lins

0.70

izons

0.69

rica

0.68

opers

0.68

atur

0.68

Activations Density 0.041%

instances of years or timeframes

references to time indicators such as "year" and "month."

No Comments

No Known Activations