INDEX

Explanations

terms related to potential consequences or significance of events or findings

oai_token-act-pair · gpt-3.5-turbo

discussions about the consequences or effects of various findings or situations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

MQ

-0.73

cop

-0.72

fighters

-0.71

Interstitial

-0.70

bows

-0.67

girls

-0.66

few

-0.65

ced

-0.64

bug

-0.64

cker

-0.63

POSITIVE LOGITS

 implications

0.97

 ramifications

0.90

romeda

0.85

 consequential

0.79

forward

0.78

uality

0.77

 thereof

0.77

ogene

0.73

 ripple

0.72

notation

0.72

Activations Density 0.055%

terms related to potential consequences or significance of events or findings

discussions about the consequences or effects of various findings or situations

No Comments

No Known Activations