INDEX

Explanations

terms related to arguments and debates

oai_token-act-pair · gpt-3.5-turbo

references to arguments or discussions

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

eco

-0.73

 livest

-0.71

cler

-0.69

ummer

-0.68

 Atomic

-0.67

 Seym

-0.67

isner

-0.65

aches

-0.64

oho

-0.63

Mos

-0.61

POSITIVE LOGITS

uments

1.12

ative

1.05

 arguments

0.99

 argument

0.89

ument

0.87

ļéĨĴ

0.85

acle

0.81

 arguing

0.77

 against

0.77

-+-+

0.76

Activations Density 0.024%

terms related to arguments and debates

references to arguments or discussions

No Comments

No Known Activations