INDEX

Explanations

phrases indicating events or actions that have occurred or been announced by a specific entity

oai_token-act-pair · gpt-3.5-turbo

the repeated use of the word "this" 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 @ 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

tops

-0.69

worms

-0.66

doms

-0.66

masters

-0.65

-0.64

Sword

-0.63

okers

-0.62

ovy

-0.62

ãĤ¹ãĥĪ

-0.60

©¶æ¥µ

-0.59

POSITIVE LOGITS

 week

1.02

 month

0.91

 year

0.90

 morning

0.86

 weekend

0.86

 latest

0.85

 afternoon

0.83

 century

0.80

 particular

0.79

decade

0.77

Activations Density 0.248%

phrases indicating events or actions that have occurred or been announced by a specific entity

the repeated use of the word "this" in various contexts

No Comments

No Known Activations