INDEX

Explanations

expressions related to gratitude and thanks

oai_token-act-pair · gpt-3.5-turbo

references to gratitude and acknowledgment 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 understatement

-0.90

illard

-0.87

arenthood

-0.75

esson

-0.73

dain

-0.70

---------

-0.68

agate

-0.66

eve

-0.64

ebus

-0.63

 misses

-0.63

POSITIVE LOGITS

 periodically

0.97

 capsules

0.87

 circular

0.81

 spherical

0.81

 predetermined

0.81

 randomly

0.77

similar

0.77

 rotating

0.75

shaped

0.74

 specially

0.73

Activations Density 1.148%

expressions related to gratitude and thanks

references to gratitude and acknowledgment in various contexts

No Comments

No Known Activations