INDEX

Explanations

phrases indicating mutual interactions or relationships between entities

oai_token-act-pair · gpt-3.5-turbo

references to interactions between individuals

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

rovers

-0.69

stones

-0.68

Bs

-0.65

hawks

-0.65

OV

-0.65

Chain

-0.64

haven

-0.64

kamp

-0.63

Jer

-0.62

mount

-0.62

POSITIVE LOGITS

 successive

1.20

 individual

0.96

 individually

0.87

 respective

0.86

 iteration

0.84

 side

0.79

THING

0.78

 succeeding

0.77

 subsequent

0.75

 element

0.75

Activations Density 0.040%

phrases indicating mutual interactions or relationships between entities

references to interactions between individuals

No Comments

No Known Activations