INDEX

Explanations

phrases related to specific lists or instructions

oai_token-act-pair · gpt-3.5-turbo

sections of text with instructions or guidelines

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hement

-0.76

 resc

-0.72

 speeding

-0.70

 showc

-0.69

 citiz

-0.69

 renewed

-0.65

 paras

-0.63

 subpoen

-0.62

 neighb

-0.62

 senior

-0.62

POSITIVE LOGITS

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

0.94

http

0.89

ÂłÂłÂłÂłÂłÂłÂłÂł

0.88

Spoiler

0.88

ÂłÂłÂłÂł

0.87

https

0.86

âĹı

0.85

oak

0.82

ALT

0.82

âĺħ

0.79

Activations Density 0.091%

phrases related to specific lists or instructions

sections of text with instructions or guidelines

No Comments

No Known Activations