INDEX

Explanations

details and levels of detail within texts

oai_token-act-pair · gpt-3.5-turbo

references to levels of detail in descriptions

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

onz

-0.84

nder

-0.77

aden

-0.72

anski

-0.72

mire

-0.70

orea

-0.69

ammy

-0.68

si

-0.67

Mos

-0.66

arte

-0.65

POSITIVE LOGITS

 detail

1.08

 details

0.84

 Detail

0.75

iveness

0.75

ariat

0.74

uring

0.73

ured

0.72

displayText

0.68

 settings

0.67

 glean

0.67

Activations Density 0.016%

details and levels of detail within texts

references to levels of detail in descriptions

No Comments

No Known Activations