INDEX

Explanations

mentions of specific organizations or groups

oai_token-act-pair · gpt-3.5-turbo

the term "und" in various contexts

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

ãĥīãĥ©ãĤ´ãĥ³

-0.90

ãĥīãĥ©

-0.82

======

-0.81

Ò

-0.80

utic

-0.79

GY

-0.76

nect

-0.74

ãĤ·ãĥ£

-0.72

âĸĪâĸĪâĸĪâĸĪ

-0.68

 toget

-0.67

POSITIVE LOGITS

erest

1.03

efined

0.98

ecided

0.90

igan

0.88

oubtedly

0.87

ocumented

0.83

lich

0.80

lings

0.79

ishly

0.77

rawn

0.77

Activations Density 0.011%

mentions of specific organizations or groups

the term "und" in various contexts

No Comments

No Known Activations