INDEX

Explanations

questions starting with "What would" or similar variations

oai_token-act-pair · gpt-3.5-turbo

conditional phrases or hypothetical situations

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

haus

-0.75

fort

-0.68

fox

-0.67

ibaba

-0.67

anie

-0.64

cule

-0.64

ledge

-0.64

belt

-0.64

hill

-0.63

skirts

-0.63

POSITIVE LOGITS

?]

0.78

 happen

0.74

 millenn

0.68

 entail

0.63

 theolog

0.61

deg

0.60

ENTS

0.59

 reconc

0.59

 corrections

0.59

?)

0.59

Activations Density 0.076%

questions starting with "What would" or similar variations

conditional phrases or hypothetical situations

No Comments

No Known Activations