INDEX

Explanations

phrases related to events happening or occurring

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "came."

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

hedon

-0.79

²¾

-0.78

illusion

-0.77

raid

-0.75

guided

-0.75

olor

-0.75

relevant

-0.74

orthodox

-0.72

rendered

-0.69

fashion

-0.69

POSITIVE LOGITS

 undone

1.14

 ashore

0.92

 forth

0.89

out

0.78

 pouring

0.78

 flooding

0.76

up

0.74

 roaring

0.74

 crashing

0.74

 forward

0.72

Activations Density 0.065%

phrases related to events happening or occurring

occurrences of the word "came."

No Comments

No Known Activations