INDEX

Explanations

words related to a positive or negative impact on specific entities or industries

oai_token-act-pair · gpt-3.5-turbo

phrases indicating significant events or outcomes

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

 indist

-0.69

 interchange

-0.67

 dens

-0.66

 pronouns

-0.65

 equivalents

-0.65

 rubbish

-0.64

 exits

-0.63

ver

-0.61

 imperson

-0.61

 curtains

-0.61

POSITIVE LOGITS

especially

0.94

particularly

0.87

 Especially

0.81

Especially

0.76

ASA

0.70

payers

0.69

 namely

0.68

utical

0.68

erning

0.67

 Particularly

0.67

Activations Density 0.591%

words related to a positive or negative impact on specific entities or industries

phrases indicating significant events or outcomes

No Comments

No Known Activations