INDEX

Explanations

words related to communication, particularly focusing on expressing or transmitting information

oai_token-act-pair · gpt-3.5-turbo

terms related to communication or expression of ideas

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

chens

-0.69

 Â§Â§

-0.66

bre

-0.65

 Nanto

-0.64

 shalt

-0.64

 Leone

-0.60

©¶æ

-0.59

Lv

-0.58

FE

-0.57

ches

-0.57

POSITIVE LOGITS

 conveyed

0.85

cha

0.83

engers

0.82

sheets

0.81

worldly

0.79

ance

0.78

atform

0.77

istor

0.74

plain

0.73

iveness

0.73

Activations Density 0.022%

words related to communication, particularly focusing on expressing or transmitting information

terms related to communication or expression of ideas

No Comments

No Known Activations