INDEX

Explanations

words related to specific locations or entities

oai_token-act-pair · gpt-3.5-turbo

mentions of familial relationships and prominent figures

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SPONSORED

-1.09

Secondly

-0.85

 Secondly

-0.82

↵Âł

-0.77

:,

-0.77

 assum

-0.76

viz

-0.74

 secondly

-0.71

âĹ¼

-0.67

 furthermore

-0.66

POSITIVE LOGITS

orneys

0.81

 Latest

0.80

 embattled

0.71

 NASCAR

0.70

attled

0.69

zens

0.68

 cybersecurity

0.67

 watchdog

0.66

nette

0.65

 widening

0.64

Activations Density 0.258%

words related to specific locations or entities

mentions of familial relationships and prominent figures

No Comments

No Known Activations