INDEX

Explanations

phrases related to financial activities and political events

oai_token-act-pair · gpt-3.5-turbo

phrases or references to organized groups or entities involved in actions or controversies

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

 sunset

-0.78

¬¼

-0.73

ishment

-0.62

xt

-0.61

earable

-0.61

IMP

-0.61

 enchantment

-0.60

ity

-0.60

FIX

-0.59

rid

-0.59

POSITIVE LOGITS

who

1.27

 whom

1.13

who

1.12

doms

1.06

 including

1.04

whose

0.94

 respectively

0.89

 albeit

0.89

 namely

0.88

particularly

0.84

Activations Density 0.447%

phrases related to financial activities and political events

phrases or references to organized groups or entities involved in actions or controversies

No Comments

No Known Activations